Mechanisms of Niche Adaption by Yersinia

Home , Yersinia enterocolitica, Yersinia entomophaga, Yersinia nurmii, Yersinia pekkanenii, Yersinia wautersii

University of Veterinary Medicine Hannover

Mechanisms of niche adaption by Yersinia

THESIS

submitted in partial fulfillment of the requirements for the degree of

Doctor rerum naturalium

(Dr. rer. nat)

awarded by the University of Veterinary Medicine Hannover

Carina Maria Schmühl

08.02.1990, Düren

Hannover, Germany 2018

Supervisor: Prof. Dr. Petra Dersch

Supervision Group: Prof. Dr. Petra Dersch (Helmholtz Centre for Infection Research, Braunschweig) Prof. Dr. Peter Valentin-Weigand (University of Veterinary Medicine Hannover) PD Dr. Simone Bergmann (Technical University Braunschweig)

1st Evaluation: Prof. Dr. Petra Dersch (Department of Molecular Infection Biology, Helmholtz Centre for Infection Research, Braunschweig) Prof. Dr. Peter Valentin-Weigand (Institute for Microbiology, University of Veterinary Medicine Hannover) PD Dr. Simone Bergmann (Institute of Microbiology, Technical University Braunschweig)

2nd Evaluation: Prof. Dr. Thilo Fuchs (Institute of Molecular Pathogenesis, Friedrich- Loeffler-Institut, Jena)

This PhD Thesis was performed at the Helmholtz Centre for Infection Research, Braunschweig, at the Department of Molecular Infection Biology.

Date of final exam: 30.10.2018

Sponsorship: Carina Maria Schmühl was supported within the Ph.D. program ’Animal and Zoonotic Infections’ by a Lichtenberg Fellowship from the Niedersächsische Ministerium für Wissenschaft und Kultur (MWK).

Parts of this thesis have been published previously:

Publications Nuss A, Beckstette M, Pimenova M, Schmühl C, Opitz W, Pisano F, Heroven AK, Dersch P. 2017. Tissue dual RNA-seq allows fast discovery of infection-specific functions and riboregulators shaping host-pathogen transcriptomes. Proc Natl Acad Sci USA. 114(5): E791-E800. doi: 10.1073/pnas.1613405114.

Schmühl C, Beckstette M, Heroven AK, Bunk B, Spröer C, McNally A, Overmann J, Dersch P. 2018. Comparative transcriptomic profiling of Yersinia enterocolitica O:3 and O:8 reveals major expression differences of fitness- and virulence-relevant genes indicating ecological separation. Nucleic Acid Res. Submitted.

Conference participations Nuss AM, Beckstette M, Schmühl C, Pisano F, Heroven AK, Reinkensmeier J, Dersch P.: Tissue Dual RNA-seq of Yersinia pseudotuberculosis and murine Peyers Patches. (Poster Presentation) 5th National Yersinia Meeting. Münster, 2016.

Schmühl C, Nuss A, Heroven AK, Beckstette M, Dersch P.: Mechanisms of the adaption and persistence of Yersinia enterocolitica. (Poster Presentation) Microbiology and Infection 2017 - 5th Joint Conference of the DGHM & VAAM. Würzburg, 2017.

Schmühl C, Beckstette M, Heroven AK, Nuss A, Dersch P.: Transcriptional landscape of Yersinia enterocolitica O:8 and O:3. (Poster Presentation) Annual Conference of the Association for General and Applied Microbiology (VAAM). Wolfsburg, 2018.

Schmühl C, Beckstette M, Heroven AK, Nuss A, Dersch P.: Transcriptional landscape of Yersinia enterocolitica O:8 and O:3. (Oral presentation) European Yersinia Conference. Jena, 2018.

Schmühl C, Beckstette M, Heroven AK, Nuss A, Dersch P.: Transcriptional landscape of Yersinia enterocolitica O:8 and O:3. (Poster Presentation) Young Microbiologist Symposium on Microbe Signaling, Organization and Pathogenesis. Belfast, 2018.

Table of Contents

TABLE OF CONTENTS I LIST OF ABBREVIATIONS I LIST OF FIGURES I LIST OF TABLES I

1 SUMMARY 1

1 ZUSAMMENFASSUNG 3

2 INTRODUCTION 5 2.1 THE GENUS YERSINIA 6 2.2 INFECTION ROUTE OF ENTEROPATHOGENIC YERSINIAE 8 2.3 SEROTYPES OF YERSINIA ENTEROCOLITICA 11 2.4 RESERVOIRS OF ENTEROPATHOGENIC YERSINIAE 14 2.5 VIRULENCE FACTORS OF ENTEROPATHOGENIC YERSINIAE 15 2.5.1 UREASE 15 2.5.2 MOTILITY 15 2.5.3 ADHESINS AND INVASINS 16 2.5.3.1 Invasin 16 2.5.3.2 Ail 18 2.5.3.3 YadA 18 2.5.4 THE PLASMID-ENCODED YSC TYPE-3-SECRETION SYSTEM 19 2.5.5 LIPOPOLYSACCHARIDES 20 2.6 Y. ENTEROCOLITICA STRAIN SPECIFIC VIRULENCE GENES 23 2.7 RNA-SEQUENCING AS A GLOBAL APPROACH TO IDENTIFY NEW REGULATORY RNAS 25 2.8 AIM OF THE STUDY 30 2.9 REFERENCES 31

3 PUBLICATION 1 47

4 PUBLICATION 2 49 4.1 ABSTRACT 50 4. 2 INTRODUCTION 50 4.3 MATERIAL AND METHODS 52 4.4 RESULTS AND DISCUSSION 59 4.4.1 COMPARATIVE RNA-SEQ OF Y. ENTEROCOLITICA O:8 AND O:3 59 4.4.2 GENOME-WIDE ANALYSIS OF TRANSCRIPTIONAL START SITES 62 4.4.3 GLOBAL ANALYSIS OF THE PROMOTER REGIONS AND ARCHITECTURE 63 4.4.4 THE REPERTOIRE OF Y. ENTEROCOLITICA NON-CODING RNAS 65 4.4.5 MONITORING OF INFECTION-RELEVANT CHANGES IN YEO:3 AND YEO:8 GENE EXPRESSION 67 4.4.6 DIFFERENTIAL EXPRESSION OF THE YSTA TOXIN GENE 76 4.5 CONCLUSIONS 79 4.6 DATA AVAILABILITY 81 4.7 SUPPLEMENTARY DATA 81 4.7.1 SUPPLEMENTARY TABLES 81

Table of Contents

4.7.2. SUPPLEMENTARY FIGURES 86 4.8 ACKNOWLEDGEMENT 92 4.9 FUNDING 92 4.10 REFERENCES 92

5 DISCUSSION 99 5.1 GLOBAL MAPPING OF TRANSCRIPTIONAL START SITES IN YERSINIA 99 5.2 THE YERSINIA REPERTOIRE OF NON-CODING RNAS 102 5.3 CHANGES IN THE GENE EXPRESSION PROFILE 105 5.4 REFERENCES 114

6 OUTLOOK 121

7 APPENDIX 123 7.1 CURRICULUM VITAE 123 7.3 ACKNOWLEDGMENT 124

List of Abbreviations

°C degree Celsius asRNA antisense RNA BHI Brain-heart infusion bp base pairs CDS coding sequence cDNA complementary DANN cfu colony forming unit Csr Carbon storage regulator DC dendritic cell DNA deoxyribonucleic acid dsDNA double-stranded DNA et al. et alii FC fold change g gram GalNac N-acetyl-galactosamine h hour(s) kb kilobasepairs kDa Kilodalton l liter LPS lipopolysaccharide mM millimolar M molar M cell Microfold cell mg milligram ml mililiter MLN mesenteric lymph node mRNA messenger RNA ncRNA non-coding RNA nt nucleotides ORF open reading frame PCR polymerase chain reaction PMN polymorphonuclear cells neutrophil PP(s) Peyer’s patch(es) pYV plasmid of Yersinia virulence qRT-PCR quantitative real-time reverse transcription-PCR RNA ribonucleic acid RNA-seq RNA-sequencing RPKM reads per kilobase transcript length per million mapped reads rRNA ribosomal RNA sRNA small regulatory RNA T3SS Type-3-Secretion-System TCA tricarboxylic acid TEX Terminator-5’-Phosphate-Dependent-Exonuclease TSS transcriptional start site UTR untranslated region YeO:3 Yersinia enterocolitica serotype O:3 YeO:8 Yersinia enterocolitica serotype O:8

III

List of figures

Figure 2.1: The phylogeny of the genus Yersinia...... 8 Figure 2.2: Invasion of enteropathogenic Yersinia into epithelial M-cells and underlying lymphatic tissues...... 10 Figure 2.3: Gene gain and gene loss in Y. enterocolitica. Y. enterocolitica strains are subdivided into different phylogoups...... 12 Figure 2.4: Schematic overview of virulence factors of enteropathogenic Yersinia expressed on the cell surface...... 16 Figure 2.5: Schematic representation of the ysc T3SS...... 20 Figure 2.6: Schematic representation of the Y. enterocolitica O:3 lipopolysaccharide...... 21 Figure 2.7: Temperature-dependent variations of lipid A in Y. enterocolitica...... 23 Figure 2.8: Model of regulation mechanisms by regulatory RNAs...... 28

Figure 4.1: Comparative RNA-seq workflow and global reports...... 61 Figure 4.2: Comparative analysis of mRNA transcriptional start sites (TSSs) of Y1 and 8081v...... 63 Figure 4.3: Identification of ncRNAs of YeO:3 Y1 and YeO:8 8081v...... 65 Figure 4.4: Comparison of the growth- and temperature-dependent regulons of YeO:3 Y1 and YeO:8 8081v...... 66 Figure 4.5: Bacterial global gene expression analysis of YeO:3 Y1 and YeO:8 8081v uncovers strain-specific metabolic and stress adaptations...... 71 Figure 4.6: Differentially regulated virulence functions between strains YeO:3 Y1 and YeO:8 8081v...... 74 Figure 4.7: Analysis of ystA expression in Y. enterocolitica...... 75 Figure 4.8: Influence of H-NS, YmoA and RovA on ystA expression of Y. enterocolitica strain Y1...... 78

Figure S4.1: Average nucleotide identity of the YeO:3 strain Y1 with other Y. enterocolitica and Y. pseudotuberculosis strains...... 86 Figure S4.2: Global identification of mRNA transcriptional start sites (TSSs)...... 87 Figure S4.3: Identification of ncRNAs of YeO:3 Y1 and YeO:8 8081v...... 88 Figure S4.4: Comparison of gene expression changes obtained by RNA-seq and real-time qRT-PCR...... 89 Figure S4.5: Gene expression analysis of stress adaption genes and regulators of YeO:3 and YeO:8...... 90 Figure S4.6: Promoter region of the ystA gene in different Y. enterocolitica strains...... 91

List of tables

Table 2.1: Y. enterocolica biogroups and serogroups (Modified from Bottone, 1999). .... 13 Table S4.1: Bacterial strains, plasmids and oligonucleotides ...... 81

1 Summary

Mechanisms of niche adaption of Yersinia Carina Schmühl

Yersinia enterocolitica and Yersinia pseudotuberculosis are zoonotic pathogens and an important cause of bacterial gastrointestinal infections. This study presents transcriptomic analysis of both species, unraveling reprogramming of the transcriptional landscape under infection-relevant conditions. Tissue dual RNA-sequencing was applied to Y. pseudotuberculosis infected mice, leading to the identification of genes upregulated during the infection process. Additionally, several ncRNAs were upregulated in vivo. Deletion of single ncRNAs did not result in reduced virulence, probably due to redundancy in ncRNA function. However, simultaneous deletion of four ncRNAs in one strain significantly impaired the colonization of the murine host. Comparative high-resolution transcriptome analysis of Y. enterocolitica serotype O:8 (YeO:8) and O:3 (YeO:3) was performed. YeO:8 is highly mouse pathogenic and well characterized, whereas YeO:3 is frequently isolated from porcine and human hosts. This study provides the first global analysis of Y. enterocolitica on a single-nucleotide resolution. The transcriptome analysis revealed a global map of transcriptional start sites (TSS) for both serotypes, identifying TSS that were conserved among or specific for the serotypes. This approach also led to the identification of conserved and strain-specific ncRNAs and regulatory RNA elements that could contribute to the differential gene expression among the two serotypes. Moreover, comparative transcriptomics revealed gene expression differences of several virulence-relevant genes, indicating certain niche-adapted properties. Among the genes being upregulated in the recent outbreak strain of YeO:3 was the enterotoxin gene ystA. The expression of ystA was shown to be independent of differences in the promoter region between YeO:8 and YeO:3; it rather depends on a regulatory factor that differs between the strains. Overall, this study provides insights into the transcriptome organization of the enteropathogenic Yersinia species and reveals gene expression differences that could lead to phenotypic variation and adaption to certain niches.

1 Summary

1 Zusammenfassung

Anpassungsmechnismen an Wirtsnischen von Yersinia Carina Schmühl

Yersinia enterocolitica und Yersinia pseudotuberculosis sind zoonotische Pathogene und eine häufige Ursache für bakterielle, gastrointestinale Infektionen. In der vorliegenden Studie wurden anhand von Transkriptomanalysen die Veränderungen im Transkriptionsprofil der beiden Spezies unter infektionsrelevanten Bedingungen gezeigt. Mittels Tissue dual RNA-sequencing von Y. pseudotuberculosis infizierten Mäusen konnten Gene identifiziert werden, die während des Infektionsprozesses induziert werden. Dazu gehörten auch einige nicht-kodierende RNAs (ncRNAs). Eine Deletion einzelner ncRNAs hatte keinen Einfluss auf die Virulenz der Yersinien im Mausmodell. Eine signifikante Reduktion der Kolonisation trat hingegen auf, wenn mehrere ncRNAs simultan deletiert wurden, was auf eine Redundanz der verschiedenen ncRNAs hindeuten könnte. Im Rahmen der Studie wurde außerdem eine vergleichende Transkriptomanalyse mit den Y. enterocolitica Serotypen O:8 (YeO:8) und O:3 (YeO:3) durchgeführt. YeO:8 zeigt eine hohe Pathogenität im Mausmodell und wurde in vielen Studien gut charakterisiert. Der am häufigsten aus Schweinen und Menschen isolierte Serotyp ist hingegen YeO:3. Die vorliegende Studie stellt die erste globale Analyse von Y. enterocolitica in Einzel-Nukleotid Auflösung dar. Diese Analyse ermöglichte die Erstellung globaler Karten von konservierten und spezifischen Transkriptionsstarts (TSS) für beide Serotypen. Außerdem konnten konservierte und Stamm-spezifische ncRNAs und regulatorische RNA Elemente identifiziert werden, welche die differentielle Genexpression beider Serotypen beinflussen könnten. Darüber hinaus konnten anhand der vergleichenden Transkriptomanalyse Unterschiede in der Genexpression verschiedener Virulenz-relevanter Gene festgestellt werden. Dies deutet auf eine Anpassung der Serotypen an verschiedene Wirtsnischen hin. So ist im YeO:3 Ausbruchsstamm unter anderem das Gen ystA induziert, welches ein Enterotoxin kodiert. Es konnte gezeigt werden, dass die differentielle Expression von ystA nicht auf Unterschiede in der Promoterregion von YeO:8 und YeO:3 zurückzuführen ist. Sie scheint eher das Ergebnis von in den beiden Serotypen unterschiedlich exprimierten regulatorischen Faktoren von ystA zu sein. Zusammengefasst liefert die vorliegende Studie Einblicke in die Organisation der Transkriptome von enteropathogenen Yersinien. Es wurden Unterschiede in der Genexpression gezeigt, welche der Grund für phänotypische Variationen und die Adaption an unterschiedliche Wirtsnichen sein könnten.

1 Zusammenfassung

2 Introduction

Zoonotic infections are a major health threat worldwide, being the dominant group of emerging infectious diseases (Jones et al., 2008). Following respiratory diseases, they are the second largest group of infectious diseases reported in Germany (Epidemiologisches Bulletin 46/2003, RKI). In 2001 and 2002, zoonoses amounted to 50 % and 60 % of all reported infections, respectively (Epidemiologisches Bulletin 46/2003, RKI). The majority of zoonoses are caused by bacteria (Jones et al., 2008). Many result in gastrointestinal diseases. The major agents of gastrointestinal infections are Salmonella, Escherichia coli (EHEC), Campylobacter and Yersinia (Epidemiologisches Bulletin 46/2003, RKI). Due to their invasive properties and their ability to colonize multiple host species, enteropathogenic Yersiniae are a suitable model organism to investigate the mechanisms underlying zoonotic infections. In the European Union, 7,017 cases of Yersiniosis were confirmed in 2011, slightly increasing to 7,202 reported cases in 2015 (Bancerz-Kisiel and Szweda, 2015; EFSA, 2015). The actual number of Yersinia infections is likely to be even higher as patients with mild symptoms might not seek treatment. These numbers make Yersiniosis the third most common bacterial zoonosis in the EU (EFSA, 2015). The EU notification rate in 2015 was 2.2 cases per 100.000 inhabitants, which was 6.8 % higher than in 2014 (EFSA, 2015). The highest notification rates were observed in north-eastern Europe, with the highest country-specific rates in Finland and Denmark (10.64 and 9.54 cases per 100.000 inhabitants, respectively) (EFSA, 2015). In Germany, 2,752 cases of Yersinia infections were reported in 2015, which is the highest total number of cases in the EU (EFSA, 2015). This number increased compared to 2014 when 2,485 cases of Yersinia infections were reported in Germany (Epidemiologisches Bulletin 20/2015, RKI), making Yersiniosis one of the most common infectious diseases in Germany (Epidemiologisches Bulletin 41/2006). Most cases of Yersiniosis (99,5%) in the EU are caused by Yersinia enterocolitica (EFSA, 2015). The knowledge about enteropathogenic Yersinia and the infections they cause has increased steadily. However, it is still important to gain a better understanding about which bacterial and host factors contribute to infection with Yersinia and how the pathogen interacts with its various host organisms. This knowledge can aid in the development of treatments directed against Yersinia infections.

2 Introduction

2.1 The genus Yersinia

Yersiniae are rod-shaped, Gram-negative, facultative anaerobes belonging to the family of Enterobacteriaceae (Smego et al., 1999). The genus Yersinia consists of 18 known species, of which three are human pathogens: Yersinia pestis, the causative agent of the plague, and the enteropathogenic species Yersinia enterocolitica and Yersinia pseudotuberculosis (Bottone, 1997; Hurst et al., 2011; Koornhof et al., 1999; Murros-Kontiainen et al., 2011a, 2011b; Savin et al., 2014; Sulakvelidze, 2000). The remaining species can be found in soil and aquatic environments. They are generally considered to be non-pathogenic, although some can be pathogenic in hosts other than mammals. Yersinia ruckerii, for example is the causative agent of the red mouth disease in salmonids and Yersinia entomophaga infects various kinds of insects (Hurst et al., 2011; Sulakvelidze, 2000). Yersinia pestis is transmitted to humans by flea bites and can spread via infected rodents, which are the primary reservoir of this pathogen (Brubaker, 1991; Wren, 2003; Zhou and Yang, 2009). In the human host, the bacteria cause severe symptoms such as fever and inflammation of the lymph nodes, which results in a painful swelling. This infection is referred to as the bubonic plague. Without treatment, the infection is usually lethal (Prentice and Rahalison, 2007). When the pathogen spreads to the lungs of an infected person, it causes the pneumonic plague. From the lung it can be transmitted to other humans via droplet infection (Wren, 2003; Zhou and Yang, 2009). Until now there have been three pandemics of the plague. The first was the Justinian plague, which started in Egypt and spread through the Middle East between the 5th and 7th century. It was followed by the second pandemic, known as the black death. It began in the 13th century with epidemics continuing into the 17th century. The modern pandemic of the plague began in 1855 in China and is still ongoing until today (Perry and Fetherston, 1997; Wren, 2003). Nowadays Y. pestis is still endemic around the world, including occasional severe outbreaks mainly in African countries and rare cases in the USA (Butler, 2013). Worldwide over 1600 people died from the plague in the first decade of the 21th century (Butler, 2013). In 2017 a major outbreak was reported in Madagascar, with over 2000 cases between August and November, including 171 deaths (WHO, 2018). Y. enterocolitica and Y. pseudotuberculosis are gastrointestinal pathogens, which are transmitted via contaminated food or water. They are able to grow at temperatures from 4°C to 42°C with optimal growth at 25°C (Bottone, 1999; Bradley et al., 1997; Brubaker, 1991). Inside the human host, Y. enterocolitica and Y. pseudotuberculosis usually cause Yersiniosis, self-limiting gut associated symptoms including invasive diarrhea, vomiting,

2 Introduction

abdominal pain, gastroenteritis and acute mesenteric lymphadenitis. Enteropathogenic Yersinia can also spread from the gastrointestinal tract and cause systemic infections of spleen and liver. Furthermore, immunological complications such as reactive arthritis and erythema nodosum syndrome can occur (Bottone, 1999; Foley and Mathews, 1984; Koornhof et al., 1999; Smego et al., 1999; Wren, 2003). Systemic infections of enteropathogenic Yersinia are rare but result in high mortality (Deacon et al., 2003). Moreover, they are able to cause septicemia in patients with hemochromatosis (Bottone, 1999). The incubation time for intestinal Yersiniosis ranges from three to seven days (Smego et al., 1999). Yersiniosis was reported as the third most common bacterial zoonotic infection in the European Union in 2015 (EFSA, 2015). In the recent years, worldwide outbreaks of enteropathogenic Yersinia have been reported in countries including Norway, New Zealand and the USA (Chakraborty et al., 2015; MacDonald et al., 2016; Williamson et al., 2016). Antibodies specific against Y. enterocolitica have been identified in more than 30% of examined people in Germany and Finland, suggesting a high-rate of non-clinical Yersiniosis in the healthy population (Mäki-Ikola et al., 1997). Notification of human cases of Yersiniosis is mandatory in most EU member states, while some countries have a voluntary notification system. Only Greece and the Netherlands have no surveillance of Yersiniosis (EFSA, 2015). Yersinia contamination of food is notifiable in 10 member states, including Germany (EFSA, 2015). The members of the genus Yersinia can be divided into 14 distinct species clusters (Fig. 2.1; McNally et al., 2016; Reuter et al., 2014). Y. pestis and Y. pseudotuberculosis are found in the same cluster, whereas Y. enterocolitica belongs to a different one, suggesting they have evolved independently of each other (Duan et al., 2014; Reuter et al., 2014). It was shown that Y. pestis emerged from Y. pseudotuberculosis shortly before the first known pandemic of the plague 1,500 to 20,000 years ago (Achtman et al., 1999; Zhou and Yang, 2009). Genome analysis revealed that they show 97% homology despite the different epidemiology and disease patterns (Chain et al., 2004). Y. enterocolitica and Y. pesudotuberculosis only share a nucleotide identity of 60% and emerged from their common ancestor 42 to 187 million years ago (Achtman et al., 1999; Chain et al., 2004; Wren, 2003). Nevertheless, Y. pestis, Y. pseudotuberculosis and Y. enterocolitica all carry the virulence plasmid pYV (called pCD1 in Y. pestis), which encodes several virulence factors and is necessary for the survival and multiplication of the bacteria inside the lymphatic tissue (Cornelis et al., 1998; Joutsen et al., 2017). The virulence plasmid of Y. enterocolitica and Y. pestis show 55%

2 Introduction

sequence homology (Portnoy and Falkow, 1981). Presumably, it was separately acquired by the pathogenic Yersiniae (Fig 2.1; Reuter et al., 2014). In addition to pYV, Y. pestis contains two further plasmids, named pMT1 and pPCP1. These are required for survival within fleas, tissue invasion and the formation of capsules (Chain et al., 2004; Perry and Fetherston, 1997; Wren, 2003).

Figure 2.1: The phylogeny of the genus Yersinia. Phylogenetic tree of the genus Yersinia based on the sequence of 84 housekeeping genes. The arrows indicate the independent acquisition of the virulence plasmid pYV (from Reuter et al., 2014). 1

Common to all three human-pathogenic species is a tropism for lymphatic tissues. Also, they are capable of resisting non-specific immune responses, in particular phagocytosis and killing by macrophages and polymorphonuclear leukocytes (PMNs) (Cornelis, 1998; Perry and Fetherston, 1997).

2.2 Infection Route of enteropathogenic Yersiniae

The enteropathogenic Yersinia species, Y. enterocolitica and Y. pseudotuberculosis, enter the host through the gastrointestinal tract after ingestion of contaminated food or water. The main sources of infection are raw or undercooked pork products, but infection can also come

1 Reuter et al., 2014 is licensed under the Creative Commons Non-Commercial Attribution license.

2 Introduction

from other meat products, vegetables and milk (Bottone, 1999; Fredriksson-Ahomaa et al., 2006; Jalava et al., 2006; Nuorti et al., 2004; EFSA, 2015). Most infections occur as single cases or in small clusters of people, typically within the same household (Rosner et al., 2010; EFSA, 2015). The initial infection phase of enteropathogenic Yersinia is characterized by the invasion and colonization of the host organism. This is accompanied by the expression of appropriate virulence factors. Immediately after entering the human body, Yersinia survives the acidic environment of the stomach by expressing a chromosomally encoded urease (Young et al., 1996). Once the pathogens reach the small intestine they are able to invade the host epithelial cells by expressing the adhesion and invasion factor invasin (Isberg et al., 1987). Invasin is expressed on the bacterial surface and binds to β1-integrins on the surface of epithelial M-cells (Sansonetti, 2002). These specialized cells are characterised by a flat morphology with short microvilli (Grützkau et al., 1990; Sansonetti, 2002). M-cells actively sample and transport antigens and microorganisms to the underlying lymphatic tissue (Kraehenbuhl and Neutra, 2000). Many enteropathogens such as Yersinia, Salmonella and Shigella employ M-cells as an entry site into the host (Fig. 2.2; Grützkau et al., 1990; Sansonetti, 2002).

Binding of invasin to β1-integrins triggers the internalization of Yersinia, which then translocate through the epithelium to reach the Peyer’s Patches (Fig 2.2; Sansonetti, 2002). Peyer’s Patches are organized lymphatic tissues that are associated with the small intestine (Grützkau et al., 1990; Pepe et al., 1995). Here, the bacteria can proliferate extracellularly and establish long-term colonization (Grützkau et al., 1990; Pepe and Miller, 1993; Sansonetti, 2002). Invasion and colonization is further affected by the expression of additional adhesins such as Ail and the O-antigen-coupled lipopolysaccharides on the bacterial surface (Skurnik and Toivanen, 1993; Uliczka et al., 2011). During the infection, Yersinia is confronted with cells of the host innate immune system within the Peyer’s Patches. The bacteria avoid phagocytosis by PMNs, macrophages and dendritic cells (DCs) through injection of Yersinia outer proteins (Yops) into the eukaryotic cells via a Type Three Secretion System (T3SS) (Durand et al., 2010; Westermark et al., 2014). It was shown that Y. enterocolitica induces apoptosis in macrophages via the T3SS (Ruckdeschel et al., 1997). Moreover, the production of reactive oxygen species by PMNs is inhibited (Spinner et al., 2010). Although they mainly replicate extracellularly, enteropathogenic Yersinia have the ability to survive inside macrophages of various hosts such as sheep, pigs, and humans (McNally et al., 2016; Moreau et al., 2010). It was suggested that they might use

2 Introduction

the activation of the autophagy pathway of macrophages for their replication (Moreau et al., 2010).

Figure 2.2: Invasion of enteropathogenic Yersinia into epithelial M-cells and underlying lymphatic tissues. Enteropathogenic Yersiniae cross the epithelial barrier through M-cells. From there, the bacteria reach the underlying Peyer’s Patches. In the Peyer’s Patches, the bacteria are confronted by the host macrophages (from Sansonetti, 2004).2

From the Peyer’s Patches, the bacteria can spread into the mesenteric lymph nodes causing mesenteric lymphadenitis and subsequently resulting in systemic infections by colonizing liver, spleen and kidney. Yersinia replicates extracellularly and causes inflammation in these organs (Heesemann et al., 1993; Sansonetti, 2002; Straley and Perry, 1995). A higher risk for systemic infection exists for Y. enterocolitica, especially in immune-compromised people as well as people with hemochromatosis (Adamkiewicz et al., 1998). Acute Yersinia infections were shown to cause long-term tissue-specific damage to infected tissues (Fonseca et al., 2015). The capacity for tolerance and protective immunity of these tissues were persistently compromised (Fonseca et al., 2015). Apart from acute infections, enteropathogenic Yersiniae are also able to establish persistent infections (Avican et al., 2015; Heine et al., 2018). This process is accompanied with a reprogramming of the

2Reprinted by permission from Springer Customer Service Centre GmbH: Springer Nature, Nature Reviews Immunology (War and peace at mucosal surfaces, Sansonetti) copyright (2004).

2 Introduction

transcriptional profile of the pathogen, including the down-regulation of the Type 3 Secretion System (Avican et al., 2015, Heine et al., 2018).

2.3 Serotypes of Yersinia enterocolitica

Based on 16S rRNA sequencing, Y. enterocolitica has been assembled into two subspecies: Y. enterocolitica subsp. enterocolitica and Y. enterocolitica subsp. palearctica (Neubauer et al., 2000). Additionally, based on the results of biochemical reactions, Y. enterocolitica is subdivided into biogroups, named 1A, 1B and 2-5. Biotype 1B contains the subspecies Y. enterocolitica subsp. enterocolitica. Y. enterocolitica subsp. palearctica includes the biotypes 1A, 2, 3, 4, 5 (Neubauer et al., 2000). Moreover, there exists a biotype 6, which is extremely rare and has only ever been isolated from wild hares (Swaminathan et al., 1982; Wuthe and Aleksić, 1997). Recently, the Y. enterocolitica biotypes were reclassified into phylogroups based on the species phylogeny (Fig. 2.3; Reuter et al., 2014; Reuter et al., 2015). Strains of the biogroup 1A are mainly considered to be non-pathogenic, although it has been shown that some of these strains can be potentially pathogenic (Falcão et al., 2006; Tennant et al., 2003). New hints suggest that members of biogroup 1A (phylogroup 1) might be able to cause reactive arthritis (Tuompo et al., 2017). The biogroups 1B, 2, 3, 4, and 5 (phylogroups 2, 3, 4, 5, and 6) are generally considered to be more pathogenic (Reuter et al., 2014). Strains of these biotypes carry the virulence plasmid pYV, which is necessary for full virulence (Bottone, 1999). Biogroups 2, 3, 4 and 5 are closely related to each other, whereas 1B is more closely related to 1A than to the other pathogenic strains (Reuter et al., 2012). The evolution of these biogroups occurred over the course of several gene-gain and gene- loss events (Fig. 2.3; McNally et al., 2016). Each biotype is further subdivided into different serotypes, based on the O-antigen serotyping (Tab. 2.1; Chester et al., 1977). It was shown that Y. enterocolitica has serotype- specific cell binding and entry characteristics (Schaake et al., 2013). Besides the bacterial serotype, the course of Y. enterocolitica infections is also influenced by certain host factors. The host-specific immune responses induced by the bacteria yield differences in survival, replication within macrophages, etc. showing that each host organism reacts differently to an infection with Y. entercolitica (Schaake et al., 2013). Human infections with Y. enterocolitica is mostly associated with the following bioserotypes: 1B/O:8, 2/O:5,27, 2/O:9, 3/O:3 and 4/O:3 (Bottone, 1999; Fredriksson-Ahomaa et al., 2006).

2 Introduction

In Europe and the USA Yersiniosis is mainly caused by the bioserotype 4/O:3 (Bottone, 1999). However, serotype O:9 is commonly isolated in Switzerland, France and the United Kingdom (Bucher et al., 2008). Reactive arthritis due to Yersinia infection is linked to a patient’s HLA type, with a strong correlation between HLA-B27 and Yersinia-induced reactive arthritis (Laitinen et al., 1977). Furthermore, the development of symptoms may be dependent on the serotype of the infecting strains with the majority of reported reactive arthritis being due to Y. enterocolitica O:3 infection (Laitenen et al., 1972).

Figure 2.3: Gene gain and gene loss in Y. enterocolitica. Y. enterocolitica strains are subdivided into different phylogoups. The phylogoups are further subdivided into serotypes based on the O- antigen. Evolution of the different serotypes is marked by the independent gain and loss of genes. Phylogroup 1 includes biogroup 1A, phylogroup 2 includes biogroup 1B, phylogroup 3 includes biogroup 4, phylogroup 4 includes biogroup 2, phylogroup 5 includes biogroup 3, phylogroup 6 includes biogroup 5 (McNally et al., 2016).3

Serotype 1B/O:8 is most intensively studied, since these bacteria are highly pathogenic in the mouse infection model. Strains belonging to biotypes 2-5 show lower pathogenicity in mice (Bottone, 1999). In food samples, the most common is biotype 1A, whereas the serotype O:9 and O:3 are predominantly isolated from animal samples (EFSA, 2015). Serotype O:3 shows a higher colonization rate in pigs than any other serotype (Robins- Browne et al., 1985; Schaake et al., 2014). In the minipig model, it was shown that O:3 is

3Reprinted by permission from Springer Customer Service Centre GmbH: Springer Nature, Nature Reviews Microbiology (‚Add, stir and reduce‘: Yersinia spp. As model bacteria for pathogen evolution, McNally A, Thomson NR, Reuter S, Wren BW), copyright (2016).

2 Introduction

able to colonize the host efficiently but induces only mild signs of inflammation. In contrast, serotype O:8 is not able to colonize pigs (Schaake et al., 2014). Accordingly, with pork being the major source of infection for humans, the bioserogroup 4/O:3 is the cause of most clinical infections in humans (Bottone, 1999; Rosner et al., 2010). Within the EU, serotype O:3 makes up 82.2% of the Y. enterocolitica strains related to human cases (EFSA, 2015).

Table 2.1: Y. enterocolica biogroups and serogroups (Modified from Bottone, 1999). Biogroup Serogroup

1A O:5; O:6,30; O:7,8; O:18; O:46

1B O:8; O:4; O:13a,13b; O:18; O:20; O:21

2 O:9; O:5,27

3 O:1,2,3; O:5,27

4 O:3

5 O:2,3

Although all Y. enterocolitica strains share virulence related genes, some differences in gene expression and gene presence occur. For example, Y. enterocolitica O:8 produces and secretes its own iron chelator, called Yersiniabactin or Yersiniophore. In contrast, serotypes O:3, O:5,27 and O:9 do not produce such a siderophore, but rely on exogenous sources for iron sequestration (Bottone, 1999). Infections with the latter serotypes usually remain confined to the gastrointestinal tract, except in the case of iron excess, which increases the possibility of a systemic infection (Bottone, 1999). Different serotypes also exist for Y. pseudotuberculosis, which are used to describe isolates from different reservoirs or differences in the lipopolysaccharide structure (Cunneen et al., 2011; Kenyon et al., 2016; Magistrali et al., 2014; Nakamura et al., 2009, 2013). However, serotypes of Y. pseudotuberculosis are not as regularly used to further distinguish the species as is done for Y. enterocolitica. A reason for this might be the fact that Y. pseudotuberculosis is less diverse within the species than Y. enterocolitica, leaving no reason for further subdivision (Fig. 2.2).

2 Introduction

2.4 Reservoirs of enteropathogenic Yersiniae

Since pigs are the major reservoir for Y. enterocolitica, pork products are the main source of infection in the European Union (EFSA, 2015). Y. enterocolitica is regularly isolated from fattening pigs with a prevalence of over 60 % (von Altrock et al., 2006; Fredriksson-Ahomaa et al., 2006). In Canada, Y. enterocolitica could be isolated from 1.8 % of tonsils of slaughtered pigs (O’Sullivan et al., 2011). It has been shown that clinical human isolates of Y. enterocolitica in Germany and Finland could not be distinguished from porcine isolates (Fredriksson-Ahomaa et al., 2006). Although Y. enterocolitica is able to colonize pigs efficiently, the infected pigs usually do not develop any symptoms of disease, supporting the role of pigs as a reservoir for human infections (Najdenski et al., 2009; Nielsen et al., 1996; Robins-Browne et al., 1985; Schaake et al., 2014). Shedding of Y. enterocolitica from infected pigs could be observed for at least 21 days, but could continue for up to 49 days (Nielsen et al., 1996). Apart from pigs, the broad host spectrum of Y. enterocolitica includes dogs, cats, sheep, cattle, goats, wild rodents, deer, foxes and wild boars (Bucher et al., 2008; Fredriksson- Ahomaa et al., 2006; Wacheck et al., 2010; EFSA, 2015). Pets fed with raw pork products might also be a source of Yersinia infections due to their close contact with humans (Fredriksson-Ahomaa et al., 2006). Small rodents such as mice, voles and shrews might also serve as a host for Y. enterocolitca. However, mostly apathogenic strains were isolated from the latter (Joutsen et al., 2017). In contrast to Y. enterocolitica, which is mostly associated with fattening pigs, Y. pseudotuberculosis is mostly associated with wildlife animals like hares, birds, deer, wild boars and rodents (Le Guern et al., 2016; Wacheck et al., 2010). Outbreaks of Yersiniosis have also been linked to raw milk, pasteurized milk, iceberg lettuce, carrots and drinking water (Ackers et al., 2000; Fukushima et al., 1988; Nuorti et al., 2004; Pärn et al., 2015; Tacket et al., 1984; Vasala et al., 2014). The enteropathogenic Yersinia species are psychotrophic, which means they can multiply in refrigerated food (Keto-Timonen et al., 2016). Therefore, it was hypothesized that prolonged cold storage of contaminated food might also be a source of infection (Williamson et al., 2016). Moreover, the bacteria are able to survive in vacuum-packed products (Hartung and Gerigk, 1991). In contrast, Y. enterocolitica is sensitive to heat. Incubation at 57 °C or 60 °C for 15 minutes or 5 minutes, respectively, is sufficient to kill the pathogens (Heim et al., 1984). Therefore, thorough cooking of the products before eating is the best way to prevent infection with Y. enterocolitica.

2 Introduction

2.5 Virulence factors of enteropathogenic Yersiniae

Virulence factors are proteins produced by pathogens, which facilitate a successful infection. They are necessary for bacteria to adapt to different host niches and promote adaption, colonization, replication and spreading of the bacteria as well as help to counteract the host immune system (Perry and Fetherston, 1997; Viboud and Bliska, 2005). Virulence factors of Yersina required in the early phase of an infection are already expressed under environmental conditions to assure the immediate penetration of the host tissues. Upon host entry, a rapid modification of virulence gene expression is essential for efficient colonization of the host and to evade the immune system. Y. enterocolitica and Y. pseudotuberculosis harbor several virulence factors encoded on the chromosome and the virulence plasmid pYV (Mikula et al., 2012).

2.5.1 Urease One virulence factor important in the early stage of an infection is the enzyme urease. Yersinia expresses urease to survive the acidic environment of the stomach (Young et al., 1996). This enzyme is induced at low pH conditions (Young et al., 1996). Urease catalyzes the hydrolysis of urea to ammonia, which neutralizes protons to counteract acid stress (Miller and Maier, 2014). Therefore, urease is the first important virulence factor needed during an infection.

2.5.2 Motility Motility enables bacteria to migrate towards nutrients and colonization niches and to avoid harmful substances. The production of flagella also contributes to the initiation of biofilm formation (Kim et al., 2008). Motility is necessary for Y. enterocolitica to reach and get in contact with host epithelial cells to initiate their uptake (Young et al., 2000). The use of motility is very energy consuming. Around 50 genes and 2% of the cell’s energy are needed for the synthesis, assembly and rotation of flagella (Macnab, 1999). The expression of flagella, and therefore motility, varies between species and serotypes. Y. pseudotuberculosis and Y. enterocolitica O:8 (YeO:8) was found to be flagellated at 25°C but not at 37°C, whereas Y. enterocolitica O:3 (YeO:3) is non-motile under in vitro conditions at both temperatures (Uliczka et al., 2011). YeO:3 isolated from the illeum was flagellated and motile, but lost its motility rapidly under in vitro growth conditions (Uliczka et al., 2011). In

2 Introduction

contrast to YeO:8, temperature is not sufficient as a stimulus to induce the expression of flagella in YeO:3 in vitro.

2.5.3 Adhesins and invasins Pathogenic bacteria need to adhere to and invade the host tissues in order to establish a successful colonization. To do so, Yersinia expresses different adhesins and invasins (Fig. 2.4; Mikula et al., 2012). The three main adhesins are invasin (InvA), the attachment and invasion locus (Ail) and the Yersinia adhesion factor A (YadA). The mechanisms of these adhesins are described in the following.

Figure 2.4: Schematic overview of virulence factors of enteropathogenic Yersinia expressed on the cell surface. YadA is depicted in green, invasin in yellow, Ail in red, LPS in light grey and the o- antigen in light purple. ECM = extracellular matrix; OM = outer membrane (modified from Mikula et al., 2012).4

2.5.3.1 Invasin Invasin is a 103 kDa outer membrane protein encoded by the invA locus on the chromosome. It consists of five globular domains, and its N-terminus is anchored into the bacterial membrane (Hamburger et al., 1999). Invasin is necessary for the colonization of the

4 Mikula et al., 2012 is licensed under the Creative Commons Non-Commercial Attribution license.

2 Introduction

Peyer’s Patches by enteropathogenic Yersinia. It binds to the β1-integrin receptors on the surface of M-cells in the gut epithelium. This binding induces the uptake of the bacteria into the cells (Koornhof et al., 1999; Pepe and Miller, 1993; Viboud and Bliska, 2005). After penetration of the epithelial M-cells, invasin is also involved in the migration into deeper organs (Heesemann et al., 2006). Non-pathogenic isolates of Y. enterocolitica were found to have only a non-functional invA gene, indicating the importance of invasin for a successful infection (Pierson and Falkow, 1990). The expression of invA is regulated in response to pH, growth phase and temperature and is positively regulated by the MarR type regulator RovA (regulator of virulence A) (Heroven et al., 2004; Revell and Miller, 2000). Similar to the expression of flagella, invasin expression is also dependent on the different serotypes of Y. enterocolitica. In YeO:8, invA is predominantly expressed at moderate temperatures between 23°C and 26°C (Nagel et al., 2001; Pepe et al., 1994; Pepe et al., 1995). However, this reduced expression at higher temperatures can be overcome at low pH (Pepe et al., 1994). Low expression levels at 37°C can be overcome by adjusting of the pH of the media to 5.5 (Nagel et al., 2001; Pepe et al., 1994). Consequently, highest levels of invasion into host cells occur when the bacteria are pre-grown at moderate temperatures (Uliczka et al., 2011). In contrast, high levels of invasin could be detected on the surfaces of YeO:3 at both temperatures, 25°C and 37°C (Uliczka et al., 2011). It was shown that in YeO:3, but not in YeO:8, an insertion element (IS1667) is present in the upstream region of invA. This insertion element provides an additional promoter for the gene, which leads to the constitutive expression of invA (Uliczka et al., 2011). Moreover, an amino acid exchange from proline to serine has occurred at position 98 between RovA of YeO:8 and YeO:3. This exchange makes RovA of YeO:3 less susceptible to proteolysis at 37°C. This leads to higher amounts of RovA and, as a result, also of invasin at 37°C in YeO:3 (Uliczka et al., 2011). However, the large amount of invasin in YeO:3 does not result in an increase in cell invasion. On the contrary, cell invasion is significantly reduced for bacteria pre-grown at 25°C, possibly due to the expression of the O-antigen of the lipopolysaccharide molecules on the cell surface. The O-antigen might create steric hindrance that prevents the interaction between invasin and surface molecules of the host cell. At 37°C the O-antigen expression is repressed, which leads to a better interaction of invasin with the host cells (Uliczka et al., 2011). This invasion pattern is unlike other Y. enterocolitica and Y. pseutotuberculosis serotypes that share the invasion pattern of Y. enterocolitica O:8 (Schaake et al., 2013; Uliczka et al., 2011).

2 Introduction

2.5.3.2 Ail Ail (attachment and invasion locus) is a chromosomally encoded outer membrane protein of 17 kDa (Miller and Falkow, 1988). It forms an eight-stranded β-barrel with four extracellular loops (Miller et al., 2001). Ail increases the adhesion to epithelial cells (Bliska and Falkow, 1992; Miller and Falkow, 1988). Known targets of Ail are laminin and heparin (Yamashita et al., 2011). In the mouse infection model, Ail was detected in the Peyer’s Patches two days after an infection with Y. enterocolitica, suggesting a role in the later stages of infection. However, virulence was not reduced in an ail mutant strain (Wachtel and Miller, 1995). Still, no ail locus is present in non-pathogenic strains (Pierson and Falkow, 1990). Ail is also involved in serum resistance by binding the complement factors H and C4bp (Biedzka-Sarek et al., 2008). This corresponds to the fact that Y. entercolitica is serum-resistant at 37°C but not at moderate temperatures (Pierson and Falkow, 1993). Ail is expressed in YeO:8 at all temperatures in exponential growth phase, but in stationary growth phase only at 37°C (Pierson and Falkow, 1993). Ail may be masked by the LPS O-antigen at moderate temperatures due its smaller size and, similar to the invasin of YeO:3, this might sterically hinder the function of Ail at these temperatures (Biedzka-Sarek et al., 2005; Kolodziejek et al., 2010). Down-regulation of O-antigen expression at 37°C allows the interaction of Ail with its target receptors as well as the complement system (Pierson and Falkow, 1993; Skurnik and Bengoechea, 2003).

2.5.3.3 YadA YadA (Yersinia adhesion factor A) is a plasmid-encoded adhesion molecule with a size of 45 kDa (Bliska et al., 1993; Heesemann et al., 2006). It adopts a lollipop structure with an N- terminal globular head domain, a C-terminal membrane anchor and a connecting coiled-coil domain (Heise and Dersch, 2006). In Y. enterocolitica YadA acts as an adhesin through binding of collagen and laminin (Heise and Dersch, 2006). YadA from Y. pseudotuberculosis is additionally able to bind fibronectin. It mediates the adhesion and invasion into deeper organs during the later phase of infection (Eitel and Dersch, 2002; Heise and Dersch, 2006). Moreover, YadA is able to induce auto-agglutination of the bacteria, which enables Yersinia to form microcolonies in lyphatic tissues (Heise and Dersch, 2006; Hoiczyk et al., 2000; Mikula et al., 2012). Furthermore, like Ail, YadA confers complement resistance through interaction with factor H (Biedzka-Sarek et al., 2008). Expression of yadA is temperature- induced at 37°C and regulated by the plasmid-encoded transcriptional regulator VirF (Eitel and Dersch, 2002). When YadA is expressed at 37°C in serotype O:8 it is so abundant that it

2 Introduction

coats the entire bacterial surface (Hoiczyk et al., 2000). In YeO:3 there seems to be a cooperative mode of action between YadA and invasin at 37°C. YadA acts as an adhesin by binding to extracellular matrix proteins enabling close contact to eukaryotic host cells. This close proximity allows the initiation of invasion by invasin. This mechanism seems to be specific for YeO:3, since the expression patterns of both proteins are different in other Y. enterocolitica serotypes (Valentin-Weigand et al., 2014).

2.5.4 The plasmid-encoded ysc Type-3-Secretion system The pathogenic Yersinia strains carry a 70 kb virulence plasmid, called pYV (plasmid of Yersinia virulence) in Y. enterocolitica and Y. pseudotuberculosis. In Y. pestis the plasmid is referred to as pCD1 (Portnoy and Falkow, 1981; Zhou and Yang, 2009). The virulence plasmid encodes several virulence factors, such as the adhesin YadA, the Yersinia outer proteins (Yops) and the ysc T3SS (Chain et al., 2004). The T3SS is generally used by extracellular bacteria to deliver bacterial proteins into the cytosol of eukaryotic cells when they are in close contact with these cells (Cornelis, 1998). Synthesis of the ysc T3SS as well as the subsequent translocation of the Yops through the secretion apparatus are induced upon host cell contact and by a temperature shift from moderate temperatures to 37°C (Pettersson et al., 1996; Rosqvist et al., 1994). The plasmid-encoded ysc T3SS is built by the Ysc proteins (Fig. 2.5). Scaffold proteins YscCDJ build rings in the inner and outer bacterial membrane. The export apparatus, consisting of YscRSTUV, and the C ring (YscQ), which are involved in substrate export, is also located in the bacterial membrane. Attached to this apparatus is the ATPase complex, formed by YscLKN, that provides the energy necessary for the protein transport through the needle structure. That needle connects the bacterial cell with the host cell membrane and is built from subunits of the proteins YscI and YscF (Cornelis, 2002; Dewoody et al., 2013). YscF is exported together with YscP, which functions as a molecular ruler to determine the length of the needle (Wagner et al., 2009). The needle tip is formed by LcrV (Mueller et al., 2005). LcrV also directs the formation of a pore in the eukaryotic cell membrane, which is composed of the transmembrane proteins YopB and YopD (Håkansson et al., 1993; Mueller et al., 2005). Besides the proteins that are integrated into the membrane, the other functional group of Yops implicates the effector proteins YopEHJMOT, which are injected into the host cell. These effector proteins act as proteases, kinases, phosphatase and GTPases. They disturb cytoskeletal dynamics, act on signaling pathways, induce apoptosis and inhibit

2 Introduction

cytokine production and phagocytosis in favor of bacterial replication and survival (Cornelis, 2002; Matsumoto and Young, 2009; Viboud and Bliska, 2005).

Figure 2.5: Schematic representation of the ysc T3SS. Schaffold proteins form the ring in the inner membrane (YscCDJ, purple). The export apparatus (YscRSTUV) is shown in orange, the ATPase complex (YscNLK) in blue. The needle consists of subunits of YscI and YscF (green). LcrV and YopBD form the needle tip and a pore in the eukaryotic membrane (red) (from Dewoody et al., 2013). 5

The translocation of Yops requires close contact with the host cell. Since the needle itself has no adhesive function, the pathogen-host-contact depends on the outer membrane proteins invasin and YadA (Autenrieth et al., 1996; Cornelis, 2002). After penetration of the intestinal lymphoid tissue, Yersinia is confronted with DCs, macrophages and PMNs. Enteropathogenic Yersiniae primarily target PMNs, but also DCs and macrophages through the injection of Yops (Westermark et al., 2014).

2.5.5 Lipopolysaccharides Lipopolysaccharides (LPSs) make up an integral part of the outer membrane of Gram- negative bacteria. They are structurally diverse, but are composed of three main parts: lipid A, the core oligosaccharide and the O-antigen (Fig. 2.6). The lipid A is anchored in the outer membrane. The core oligosaccharide is made of sugar residues and is often divided into an inner and an outer core. The O-antigen is a long saccharide polymer that contains

5Dewoody et al., 2013 is licensed under the Creative Commons Non-Commercial Attribution license.

2 Introduction

repeat units made of 2-8 sugar residues each. The O-antigen is exposed to the cell surroundings.

Figure 2.6: Schematic representation of the Y. enterocolitica O:3 lipopolysaccharide (LPS). LPS is composed of three main parts: lipid A, which is anchored into the bacterial membrane; the core oligosaccharide, which is composed of sugar residues and divided further into the inner and outer core; the O-antigen, an oligosaccharide chain exposed to the cell surroundings. GlcN = glucosamine; Kdo = 3-deoxy-D-manno-2-octulosonic acid; Hep = Glycero-D-mannoheptose; FuNa = N-acetyl- fucosamine; GaNa = N-acetylgalactosamine; Gal = galactose; Glc = Glucose; 6dA = 6-deoxy-L-altrose (from Sirisena and Skurnik, 2003).6

Three basic forms of LPS exist, which are termed rough (R), semi-rough (SR) and smooth (S) (Białas et al., 2012). Smooth LPS contains all three of the LPS’s main parts. This form is present in most Enterobacteriaceae. The semi-rough LPS carries only one O-antigen unit and there is no O-antigen present in the rough LPS (Białas et al., 2012). In Y. enterocolitica it has been shown that the complete LPS is important for full virulence of the bacteria. Especially the O-antigen is an important virulence factor. Mutants expressing R-LPS and SR-LPS were shown to be strikingly less virulent than the strains expressing S-LPS. The mutants are reduced in their ability to colonize the Peyer’s Patches and the spleen (Bengoechea et al., 2004; al-Hendy et al., 1992; Zhang et al., 1997).

6Permitted for use in a dissertation / thesis by John Wiley and Sons (Mutations in the genes for synthesis of the outer core region of the lipopolysaccharide of Yersinia enterocolitica O:3, Sirisena DM Skurnik M) copyright (2013).

2 Introduction

Several additional biological roles have been attributed to the O-antigen. It was shown that it functions as a bacteriophage receptor, enhances the effects of adhesion factors and plays a role in the microbial resistance to the host complement system (Biedzka-Sarek et al., 2008; Erridge et al., 2002). Additionally, it has been shown that the outer core plays a role in the outer membrane’s permeability to hydrophobic agents and resistance to antimicrobial peptides. In contrast to wild-type bacteria, mutants lacking the outer core were not able to kill infected mice (Skurnik et al., 1999). The LPS inner core has been described to function as a phage receptor (Leon-Velarde et al., 2016). Lipid A, as well as the O-antigen, is important for the resistance against antimicrobial peptides (Reinés et al., 2012). It was demonstrated that Y. enterocolitica is more susceptible to antimicrobial peptides when grown at 37°C compared to moderate temperatures, and this effect is linked to temperature-dependent modifications of lipid A (Bengoechea et al., 1996; Reinés et al., 2012). In general, growth temperature has a significant impact on LPS modification (Rebeil et al., 2004) (Fig. 2.7;). When grown at moderate temperatures, all human-pathogenic Yersinia species contain hexa-acylated lipid A (Rebeil et al., 2004). It has been shown that this form of the LPS of Y. pestis is recognized by the host Toll-like receptor 4 (TLR-4) and induces the expression of proinflammatory cytokines. When expressed at 37°C, the Y. pestis LPS does not contain a hexa-acetylated lipid A and therefore does not induce the pro-inflammatory host response (Matsuura et al., 2010). When grown at 37°C Y. pseudotuberculosis expresses a heterogeneous population of lipid A, whereas Y. enterocolitica expresses a tetra-acylated lipid A, which is also a weak TLR-4 agonist and does not induce an immune response (Rebeil et al., 2004). Differences in LPS structure are also observed between different serotypes. The inner core of YeO:8 LPS is similar to the inner core of YeO:3 LPS, but in YeO:8 the outer core is missing. In the genome of YeO:3, the clusters involved in the synthesis of the LPS outer core and O- antigen are organized differently compared to other bioserotypes (Batzilla et al., 2011; Skurnik and Bengoechea, 2003). The LPS of YeO:3 shows a unique structure because both, the O-antigen and the branched outer core, are bound to the inner core. In other strains, the O-antigen is linked to the outer core (Sirisena and Skurnik, 2003; Skurnik et al., 1999) (Fig. 2.6). Maximum expression of the YeO:3 O-antigen is induced at moderate temperatures, while at 37°C, only low amounts are detectable (Skurnik and Bengoechea, 2003). It has been shown that an individual LPS molecule carries either the outer core or the O-antigen linked to the inner core, but not both (Pinta et al., 2012). In addition to the O-antigen, enterobacterial common antigen (ECA) can be linked to YeO:3 LPS (Muszyński et al., 2013). The expression

2 Introduction

of ECA is also temperature-regulated. The ECA and O-antigen can coexists on the same molecule (Muszyński et al., 2013).

Figure 2.7: Temperature-dependent variations of lipid A in Y. enterocolitica. When grown at 26°C, Y. enterocolitica expresses hexa-acylated lipid A, which is also common to other Yersinia species and Enterobacteriaceae. At 37°C, lipid A expressed by Y. enterocolitica is tetra-acylated (from Tsolis et al., 2008). 7

2.6 Y. enterocolitica strain specific virulence genes

The Y. enterocolitica serotypes O:8 and O:3 do not only show differences in gene expression patterns, as described for the adhesion factors and LPS, but also show genetic differences. In a direct comparison between YeO:8 and YeO:3, there are genes specific for each of the two serotypes. Besides the plasmid encoded T3SS, Y. enterocolitica harbors an additional T3SS encoded on the chromosome. In YeO:8 the ysa T3SS is located in a region on the chromosome referred to as the plasticity zone, which contains several genes associated with virulence (Thomson et al., 2006). For Y. enterocolitica, the plasticity zone, and therefore the ysa T3SS, is unique to biogroup 1B (Foultier et al., 2002; Haller et al., 2000; Thomson et al., 2006; Young and Young, 2002). Some proteins are exported by the ysa T3SS as well as by the ysc

7 Reprinted by permission from Spinger Customer Service Centre GmbH: Springer Nature, Nature Reviews Microbiology (From bench to bedside: stealth of enteroinvasive pathogens, Tsolis RM Young GM Solnick JV Bäumler AJ) copyright (2008).

2 Introduction

T3SS (Foultier et al., 2003; Young and Young, 2002). Eight additional translocated effectors have been described that are not targeted by the ysc T3SS but are exclusive for the ysa T3SS (Foultier et al., 2003; Matsumoto and Young, 2006). Each of these effectors is necessary for full virulence in the mouse model (Matsumoto and Young, 2006). The genes encoding these effectors are not located in clusters but are dispersed throughout the chromosome (Matsumoto and Young, 2006). The particular functions of these single effectors remain to be investigated. The ysa T3SS is important for the colonization of the small intestine in early phases of infection and for overcoming the host immune system in the gastrointestinal tissues (Venecia and Young, 2005). A ysa mutant strain is significantly less virulent than the wild-type strain (Haller et al., 2000). However, in the systemic phase of an infection the ysa T3SS has no effect (Venecia and Young, 2005). For full virulence of YeO:8, both the ysc T3SS and the ysa T3SS are necessary (Venecia and Young, 2005). The T3SS specific for Y. enterocolitica serotype O:3 is referred to as ysp T3SS and is also encoded on its chromosome. The ysp T3SS itself has not been thoroughly investigated, but it is highly homologous to the SPI-2 T3SS of Salmonella (Batzilla et al., 2011). Salmonalla mutants lacking the SPI-2 system are able to colonize the Peyer’s Patches, but not the mesenteric lymph nodes, liver or spleen (Marcus et al., 2000). It has also been shown that SPI-2 is important for systemic infections and intracellular survival of Salmonella (Hansen- Wester and Hensel, 2001). In contrast to Salmonella, Y. enterocolitica does not live intracellularly. Therefore, the ysp T3SS might have another, yet unknown function. Another important virulence factor for YeO:8 is the siderophore yersiniabactin (Pelludat et al., 1998). The yersiniabactin locus comprises 11 genes that are divided into three functional groups: yersiniabactin biosynthesis, transport into the bacterial cell and regulation of expression (Carniel, 2001). The locus has been shown to be regulated by the iron-sensitive repressor Fur (Carniel, 2001). Although yersiniabactin is mainly required for the acquisition of Fe3+, it was also shown to be involved in the uptake of zinc (Bobrov et al., 2014; Forman et al., 2010; Rakin et al., 2012). The system is induced in the presence of oxygen (Forman et al., 2010; Rakin et al., 2012). Specific for YeO:3 is an operon encoding for a phosphotransferase system (PTS) that is necessary for the uptake of N-acetyl-galactosamine (GalNac or Aga). Since the porcine intestine is rich in N-acetyl-galactosamine-containing mucin, the presence of this operon might support the survival of this serotype in the porcine host (Batzilla et al., 2011). Studies characterizing this system in Yersinia still need to be performed, but in E. coli the GalNac- specific phosphotransferase system has been studied. E. coli has an operon consisting of 13

2 Introduction

genes, which encode the proteins necessary for the utilization of both, GalNac and D- galactosamine (Gam) (Brinkkötter et al., 2000). The system is controlled by the repressor AgaR, which was shown to respond to the amount of available GalNac (Brinkkötter et al., 2000; Hu et al., 2013; Ray and Larson, 2004). An additional gene present in YeO:3, but absent in YeO:8, is the large gene rtxA. The RtxA protein belongs to the so-called MARTX toxins (multifunctional autoprocessing repeats-in- toxin), a toxin family of large proteins (between 3000 and 5300 amino acid residues) that are encoded in several Gram-negative bacteria (Satchell, 2007; Zhou and Yang, 2009). Characteristic for MARTX proteins is a series of glycine-rich repeats (Satchell, 2007). The best-studied example of an MARTX toxin is RtxA from Vibrio cholerae (referred to as

MARTXVC). This protein consists of more than 4500 amino acid residues and interacts with the actin of several cell types (Linhartová et al., 2010; Satchell, 2007). In Y. enterocolitica, rtxA is located in an operon with rtxH and rtxC. The function of rtxH is unknown. The putative acyltransferase RtxC contributes to toxicity in vivo (Liu et al., 2007; Satchell, 2011). In Vibrio vulnificus it was shown that rtxHCA is transcribed from a single promotor (Park et al., 2012). The expression of rtxA is induced after host cell contact (Kim et al., 2008). Moreover, the expression of the rtxHCA operon has been observed in a growth phase-dependent manner (Park et al., 2012). Studies of RtxA in V. vulnificus have revealed that it is essential for pathogenesis, and that it induces cyctoskeletal rearrangement in eukaryotic cells (Kim et al., 2008; Ziolo et al., 2014). MARTX family members are exported through the bacterial membrane via an atypical type-I-secretion-system (Boardman and Satchell, 2004; Linhartová et al., 2010). Genes for the corresponding export apparatus are disrupted in YeO:3 (Batzilla et al., 2011). It is possible that the protein is exported via a different system. The functionality of RtxA in Y. enterocolitica remains to be further investigated.

2.7 RNA-sequencing as a global approach to identify new regulatory RNAs

In recent years, RNA-sequencing (RNA-seq) has become a major tool to investigate transcriptional changes in bacteria under various conditions in vitro and in vivo. Transcriptome analysis using RNA-seq allows the determination of global changes in gene- expression patterns. Since the detection of transcripts is annotation-independent, both known and previously unknown targets can be identified using this method. The transcripts can be mapped to the genomic sequence with nucleotide resolution, enabling the identification of transcriptional start sites (TSS), operon structures and untranslated regions

2 Introduction

of mRNAs (UTRs). To date, RNA-seq has been used to detect the transcriptional landscape of many organisms, including various human pathogens such as Borrelia burgdorferi, Campylobacter jejuni, Pseudomonas aeruginosa and Vibrio cholerae (Arnold et al., 2016; Butcher and Stintzi, 2013; Dötsch et al., 2012; Mandlik et al., 2011). Even bacteria that are difficult to culture such as Chlamydia pneumonia can be analyzed using this approach (Albrecht et al., 2011). RNA-seq cannot only be used to detect the expression levels of protein-encoding genes, but it can also serve as a global approach to identify small regulatory RNAs (sRNAs). It is possible to predict sRNAs based on their genomic sequence, for example based on the presence of putative Rho-independent terminators (Xia et al., 2012). But while this approach is purely hypothetical, RNA-seq results reveal targets that are actually transcribed under the investigated conditions. RNA-seq has already been proven to be a useful tool for the identification of sRNAs. For example, in Y. pestis 104 sRNAs could be found using transcriptome analysis (Yan et al., 2013). In the last years, regulatory RNAs have been identified by means of RNA-seq in several species including Bacillus subtilis, E. coli, Helicobacter pylori, P. aeroginosa, Streptomyces coelicolor, V. cholare, Y. pestis and Y. pseudotuberculosis (Gómez-Lozano et al., 2012; Irnov et al., 2010; Koo et al., 2011; Mandlik et al., 2011; Raghavan et al., 2011; Sharma et al., 2010; Vockenhuber et al., 2011; Yan et al., 2013). Regulatory RNAs are transcripts with a size typically between 50 – 300 nts (Storz et al., 2011). Longer transcripts of sizes between 700 and 3500 nts have also been described (Steglich et al., 2008). Such sRNAs are transcribed but not translated. They regulate gene expression at the post-transcriptional level by forming an sRNA-mRNA duplex with their target mRNA. This interaction can have a positive or negative influence on target gene expression (Nitzan et al., 2017; Papenfort and Vanderpool, 2015). Interactions of sRNAs with mRNAs are important for fast and tight regulation to ensure an immediate physiological response within the bacterial cell to changes in environmental conditions. Computational analysis showed that regulation by sRNAs is faster than by transcription factors (Shimoni et al., 2007). In recent years, sRNAs have been recognized as important post-transcriptional regulators in many pathogenic species such as Pseudomonas, Salmonella, Vibrio and Yersinia (Mandlik et al., 2011; Nuss et al., 2017; Sonnleitner and Haas, 2011; Vogel, 2009). They have been connected to central roles in a variety of cellular functions like the stress response (Gottesman et al., 2006; Nuss et al., 2017), iron homeostasis (Butcher and Stintzi, 2013;

2 Introduction

Salvail and Massé, 2012), sugar metabolism (Urban and Vogel, 2008), quorum sensing (Tsai and Winans, 2011), the response to oxidative stress (Calderón et al., 2014) and the acid adaptation response (Xia et al., 2012). They are also part of regulatory networks controlling antibiotic resistance (Dersch et al., 2017; Kim et al., 2015). Regulatory RNAs can affect the expression of their target genes via different molecular mechanisms (Fig. 2.8), namely • transcription interference, • alteration of transcript stability, • translation initiation, • translation repression • alteration of protein activity. Regulation of transcription by regulatory RNAs affects the formation of regulatory structures. Binding of an sRNA has been shown to promote premature termination of transcription (Giangrossi et al., 2010; Gong et al., 2011). Other RNA molecules have a contradictory effect and prevent premature termination of transcription (Sedlyarova et al., 2016). Duplex formation between sRNA and mRNA might also affect the stability of the transcripts, resulting in the degradation of both RNAs as double-stranded RNA (dsRNA) is frequently degraded by RNase E (Massé et al., 2003; Prévost et al., 2007; Waters and Storz, 2009). On the other hand, duplex formation can also result in enhanced stability, which increases the translation rate (Dadzie et al., 2013). It is also possible that base pairing of the sRNA prevents the formation of an inhibitory secondary structure and exposes the ribosomal binding site (RBS) (Balbontín et al., 2016; Prévost et al., 2007; Urban and Vogel, 2008). Moreover, translation can be activated by preventing the formation of a translation-inhibiting loop structure (Hammer and Bassler, 2007). In contrast, binding of sRNAs can prevent ribosome binding to the mRNA, thus inhibiting the translation of the target transcript (Kawano et al., 2007; Sharma et al., 2007; Song et al., 2008).

2 Introduction

Figure 2.8: Model of regulation mechanisms by regulatory RNAs. Regulatory RNAs can positively or negatively influence the expression of their target mRNAs (from Nitzan et al., 2017). 8 sRNAs do not only interact with mRNA molecules, they can also interact with proteins and regulate their activity (Storz et al., 2011). The best-described example is the carbon storage regulator (Csr) system. The RNA molecules CsrB and CsrC have multiple binding sites for the global regulator CsrA. Hence, they are able to sequester CsrA from its target mRNAs, altering its activity (Liu et al., 1997; Weilbacher et al., 2003). Regulatory RNAs can bind at different locations within their target transcripts. Many sRNAs base pair close to the RBS (Gong et al., 2011; Prévost et al., 2007). However, they might also bind to other regions of the 5’-UTR (Obana et al., 2010; Pappesch et al., 2017; Sharma et al., 2007; Vecerek et al., 2007), in the coding sequence of the mRNA (Fröhlich et al., 2012) or in the 3’-UTR (Opdyke et al., 2004). Regulatory RNAs are usually differentiated by their chromosomal orientation towards their target genes. They are generally subdivided into three groups: regulatory elements in the 5’- UTR of an mRNA, cis-encoded antisense RNAs and trans-encoded RNAs.

8 Permitted for use in a dissertation / thesis by Annual Reviews (Integration of Bacterial Small RNAs in Regulatory Networks, Nitzan M Rehani R Margalit H), copyright (2017).

2 Introduction

Sensory elements in the 5’-UTR of mRNAs regulate the expression of the downstream- encoded genes by structural alterations as a reaction to different environmental signals, thereby masking or revealing the RBS, the Shine-Dalgarno (SD) region or the translational start codon (Narberhaus et al., 2006; Oliva et al., 2015). Those elements are referred to as riboswitches if they respond to metabolite binding. This leads to a feedback control mechanism if this metabolite is also the final product of the pathways regulated by the riboswitch. In case the induced change in structure is due to temperature, the regulating element is called an RNA thermometer (Narberhaus et al., 2006; Oliva et al., 2015; Waters and Storz, 2009). Such sensory elements cannot be detected using RNA-seq because they are part of the mRNA transcript. In contrast, cis-encoded antisense RNAs and trans-encoded RNAs are distinct transcripts and, therefore, detectable by RNA-seq. Cis-encoded antisense RNAs are encoded at the same genetic location, but on the strand opposite to their target RNA. Therefore, they have perfect complementarity to the RNA they act upon (Storz et al., 2005). Antisense RNAs can be complementary to the 3’-UTR, the coding sequence or the 5’-UTR of their target mRNA (Georg and Hess, 2011). Such RNAs usually target the specific transcript they are antisense-encoded to (Saberi et al., 2016). Nevertheless, two genes encoded on the same strand might also be connected via an antisense RNA if they have overlapping UTRs. Antisense RNAs mediate a variety of physiological effects by binding to their target transcripts (Georg and Hess, 2011). Trans-encoded RNAs are usually located at a different position in the genome than their target mRNAs (Storz et al., 2005). Generally, each trans-encoded RNA has multiple target mRNAs and might even form their own regulon (Storz et al., 2011; Waters and Storz, 2009). That is because they show only partial complementarity and therefore base pair with mismatching positions (Storz et al., 2005, 2011; Waters and Storz, 2009). The chaperone Hfq is essential for the activity of many trans-encoded RNAs in bacteria (Oliva et al., 2015; Waters and Storz, 2009). It facilitates the interaction between mRNA and sRNA. Moreover, it protects sRNAs from cleavage by ribonucleases or induces the cleavage of sRNA-mRNA duplexes (Oliva et al., 2015). However, not all trans-encoded sRNAs are dependent on Hfq for their activity (Oliva et al., 2015).

2 Introduction

2.8 Aim of the study

During their life cycle, enteropathogenic Yersiniae are exposed to different environments inside and outside host organisms. Therefore, they have to adapt their gene expression depending on the condition in order to persist. The overall goal of this study was to investigate the influence of environmental conditions on the transcriptional landscape of enteropathogenic Yersinia species. To address this question, RNA-seq should be applied. This approach allows the investigation of global changes in gene expression in response to environmental signals as well as the role of non-coding RNAs (ncRNAs) for Yersinia virulence. For Y. pseudotuberculosis this approach has been used in a previous study (Nuss et al., 2015). This has gained a deep insight into the gene expression profile and has lead to the discovery of new ncRNAs. Here, tissue-dual RNA-sequencing of Y. pseudotuberculosis in mouse Peyer’s Patches was applied to investigate changes between in vitro and in vivo conditions. One aim of this study was the analysis of a target set of genes and ncRNAs that were found to be upregulated in vivo. These targets should be investigated further to identify their role during the infection process. For this purpose, deletion mutants should be created and analyzed. Another aim of this study was to compare the transcription profile of two Y. enterocolitica serotypes in response to temperature and nutrient availability. Most cases of Yersiniosis in the EU and the USA are caused the bioserotype 4/O:3 (YeO:3) (Bottone, 1999; Bucher et al., 2008; Rosner et al., 2010). However, the bioserotype 1B/O:8 (YeO:8), highly mouse virulent and commonly used in laboratory experiments, is responsible for only 1% of the Yersiniosis cases in Germany (Rosner et al., 2010). Previous studies have shown that both, YeO:3 and YeO:8 are able to colonize mice. But while the YeO:3 infected mice stay clinically healthy, the ones infected with YeO:8 show severe pathological changes (Schaake et al., 2014). On the other hand, YeO:3 was shown to be able to colonize pigs, in contrast to YeO:8 (Schaake et al., 2014). This work aims to investigate the differences of both serotypes on the transcriptional level to gain insights into the mechanisms underlying the different colonization phenotypes. To address this, RNA-seq of Y. enterocolitica at different in vitro conditions should be established. The results should be used to analyze changes in the transcriptional profile on a global level. Gene expression and the presence of ncRNAs should be investigated and a global map of transcriptional start sites for Y. enterocolitica should be created.

2 Introduction

2.9 References

Achtman, M., Zurth, K., Morelli, G., Torrea, G., Guiyoule, A., and Carniel, E. (1999). Yersinia pestis, the cause of plague, is a recently emerged clone of Yersinia pseudotuberculosis. Proc. Natl. Acad. Sci. U.S.A. 96, 14043–14048.

Ackers, M.L., Schoenfeld, S., Markman, J., Smith, M.G., Nicholson, M.A., DeWitt, W., Cameron, D.N., Griffin, P.M., and Slutsker, L. (2000). An outbreak of Yersinia enterocolitica O:8 infections associated with pasteurized milk. J. Infect. Dis. 181, 1834–1837.

Adamkiewicz, T.V., Berkovitch, M., Krishnan, C., Polsinelli, C., Kermack, D., and Olivieri, N.F. (1998). Infection due to Yersinia enterocolitica in a series of patients with beta-thalassemia: incidence and predisposing factors. Clin. Infect. Dis. 27, 1362–1366.

Albrecht, M., Sharma, C.M., Dittrich, M.T., Müller, T., Reinhardt, R., Vogel, J., and Rudel, T. (2011). The transcriptional landscape of Chlamydia pneumoniae. Genome Biol. 12, R98. von Altrock, A., Louis, A.L., Rösler, U., Alter, T., Beyerbach, M., Kreienbrocks, L., and Waldmann, K.-H. (2006). [The bacteriological and serological prevalence of Campylobacter spp. and Yersinia enterocolitica in fattening pig herds in Lower Saxony]. Berl. Munch. Tierarztl. Wochenschr. 119, 391–399.

Arnold, W.K., Savage, C.R., Brissette, C.A., Seshu, J., Livny, J., and Stevenson, B. (2016). RNA- Seq of Borrelia burgdorferi in Multiple Phases of Growth Reveals Insights into the Dynamics of Gene Expression, Transcriptome Architecture, and Noncoding RNAs. PLoS ONE 11, e0164165.

Autenrieth, I.B., Kempf, V., Sprinz, T., Preger, S., and Schnell, A. (1996). Defense mechanisms in Peyer’s patches and mesenteric lymph nodes against Yersinia enterocolitica involve integrins and cytokines. Infect. Immun. 64, 1357–1368.

Avican, K., Fahlgren, A., Huss, M., Heroven, A.K., Beckstette, M., Dersch, P., and Fällman, M. (2015). Reprogramming of Yersinia from virulent to persistent mode revealed by complex in vivo RNA- seq analysis. PLoS Pathog. 11, e1004600.

Balbontín, R., Villagra, N., Pardos de la Gándara, M., Mora, G., Figueroa-Bossi, N., and Bossi, L. (2016). Expression of IroN, the salmochelin siderophore receptor, requires mRNA activation by RyhB small RNA homologues. Mol. Microbiol. 100, 139–155.

Bancerz-Kisiel, A., and Szweda, W. (2015). Yersiniosis - a zoonotic foodborne disease of relevance to public health. Ann Agric Environ Med 22, 397–402.

Batzilla, J., Antonenka, U., Höper, D., Heesemann, J., and Rakin, A. (2011). Yersinia enterocolitica palearctica serobiotype O:3/4--a successful group of emerging zoonotic pathogens. BMC Genomics 12, 348.

Bengoechea, J.A., Díaz, R., and Moriyón, I. (1996). Outer membrane differences between pathogenic and environmental Yersinia enterocolitica biogroups probed with hydrophobic permeants and polycationic peptides. Infect. Immun. 64, 4891–4899.

2 Introduction

Bengoechea, J.A., Najdenski, H., and Skurnik, M. (2004). Lipopolysaccharide O antigen status of Yersinia enterocolitica O:8 is essential for virulence and absence of O antigen affects the expression of other Yersinia virulence factors. Mol. Microbiol. 52, 451–469.

Białas, N., Kasperkiewicz, K., Radziejewska-Lebrecht, J., and Skurnik, M. (2012). Bacterial cell surface structures in Yersinia enterocolitica. Arch. Immunol. Ther. Exp. (Warsz.) 60, 199–209.

Biedzka-Sarek, M., Venho, R., and Skurnik, M. (2005). Role of YadA, Ail, and Lipopolysaccharide in Serum Resistance of Yersinia enterocolitica Serotype O:3. Infect. Immun. 73, 2232–2244.

Biedzka-Sarek, M., Salmenlinna, S., Gruber, M., Lupas, A.N., Meri, S., and Skurnik, M. (2008). Functional mapping of YadA- and Ail-mediated binding of human factor H to Yersinia enterocolitica serotype O:3. Infect. Immun. 76, 5016–5027.

Bliska, J.B., and Falkow, S. (1992). Bacterial resistance to complement killing mediated by the Ail protein of Yersinia enterocolitica. Proc. Natl. Acad. Sci. U.S.A. 89, 3561–3565.

Bliska, J.B., Copass, M.C., and Falkow, S. (1993).The Yersinia pseudotuberculosis adhesin YadA mediates intimate bacterial attachment to and entry into HEp-2 cells. Infect. Immun.61,3914–3921.

Boardman, B.K., and Satchell, K.J.F. (2004). Vibrio cholerae strains with mutations in an atypical type I secretion system accumulate RTX toxin intracellularly. J. Bacteriol. 186, 8137–8143.

Bobrov, A.G., Kirillina, O., Fetherston, J.D., Miller, M.C., Burlison, J.A., and Perry, R.D. (2014). The Yersinia pestis siderophore, yersiniabactin, and the ZnuABC system both contribute to zinc acquisition and the development of lethal septicaemic plague in mice. Mol. Microbiol. 93, 759–775.

Bottone, E.J. (1997). Yersinia enterocolitica: the charisma continues. Clin. Microbiol.Rev.10,257–276.

Bottone, E.J. (1999). Yersinia enterocolitica: overview and epidemiologic correlates. Microbes Infect. 1, 323–333.

Bradley, R.M., Gander, R.M., Patel, S.K., and Kaplan, H.S. (1997). Inhibitory effect of 0 degree C storage on the proliferation of Yersinia enterocolitica in donated blood. Transfusion 37, 691–695.

Brinkkötter, A., Klöss, H., Alpert, C., and Lengeler, J.W. (2000). Pathways for the utilization of N- acetyl-galactosamine and galactosamine in Escherichia coli. Mol. Microbiol. 37, 125–135.

Brubaker, R.R. (1991). Factors promoting acute and chronic diseases caused by yersiniae. Clin. Microbiol. Rev. 4, 309–324.

Bucher, M., Meyer, C., Grötzbach, B., Wacheck, S., Stolle, A., and Fredriksson-Ahomaa, M. (2008). Epidemiological data on pathogenic Yersinia enterocolitica in Southern Germany during 2000- 2006. Foodborne Pathog. Dis. 5, 273–280.

Butcher, J., and Stintzi, A. (2013). The transcriptional landscape of Campylobacter jejuni under iron replete and iron limited growth conditions. PLoS ONE 8, e79475.

Butler, T. (2013). Plague gives surprises in the first decade of the 21st century in the United States and worldwide. Am. J. Trop. Med. Hyg. 89, 788–793.

2 Introduction

Calderón, I.L., Morales, E.H., Collao, B., Calderón, P.F., Chahuán, C.A., Acuña, L.G., Gil, F., and Saavedra, C.P. (2014). Role of Salmonella Typhimurium small RNAs RyhB-1 and RyhB-2 in the oxidative stress response. Res. Microbiol. 165, 30–40.

Carniel, E. (2001). The Yersinia high-pathogenicity island: an iron-uptake island. Microbes Infect. 3, 561–569.

Chain, P.S.G., Carniel, E., Larimer, F.W., Lamerdin, J., Stoutland, P.O., Regala, W.M., Georgescu, A.M., Vergez, L.M., Land, M.L., Motin, V.L., et al. (2004). Insights into the evolution of Yersinia pestis through whole-genome comparison with Yersinia pseudotuberculosis. Proc. Natl. Acad. Sci. U.S.A. 101, 13826–13831.

Chakraborty, A., Komatsu, K., Roberts, M., Collins, J., Beggs, J., Turabelidze, G., Safranek, T., Maillard, J.-M., Bell, L.J., Young, D., et al. (2015). The descriptive epidemiology of yersiniosis: a multistate study, 2005-2011. Public Health Rep 130, 269–277.

Chester, B., Stotzky, G., Bottone, E.J., Malowany, M.S., and Allerhand, J. (1977). Yersinia enterocolitica: biochemical, serological, and gas-liquid chromatographic characterization of rhamnose-, raffinose-, melibiose-, and citrate-utilizing strains. J. Clin. Microbiol. 6, 461–468.

Cornelis, G.R. (1998). The Yersinia deadly kiss. J. Bacteriol. 180, 5495–5504.

Cornelis, G.R. (2002). Yersinia type III secretion: send in the effectors. J. Cell Biol. 158, 401–408.

Cornelis, G.R., Boland, A., Boyd, A.P., Geuijen, C., Iriarte, M., Neyt, C., Sory, M.P., and Stainier, I. (1998).The virulence plasmid of Yersinia, an antihost genome.Microbiol.Mol.Biol.Rev.62,1315–1352.

Cunneen, M.M., Pacinelli, E., Song, W.C., and Reeves, P.R. (2011). Genetic analysis of the O- antigen gene clusters of Yersinia pseudotuberculosis O:6 and O:7. Glycobiology 21, 1140–1146.

Dadzie, I., Xu, S., Ni, B., Zhang, X., Zhang, H., Sheng, X., Xu, H., and Huang, X. (2013). Identification and characterization of a cis-encoded antisense RNA associated with the replication process of Salmonella enterica serovar Typhi. PLoS ONE 8, e61308.

Deacon, A.G., Hay, A., and Duncan, J. (2003). Septicemia due to Yersinia pseudotuberculosis--a case report. Clin. Microbiol. Infect. 9, 1118–1119.

Dersch, P., Khan, M.A., Mühlen, S., and Görke, B. (2017). Roles of Regulatory RNAs for Antibiotic Resistance in Bacteria and Their Potential Value as Novel Drug Targets. Front Microbiol 8, 803.

Dewoody, R.S., Merritt, P.M., and Marketon, M.M. (2013). Regulation of the Yersinia type III secretion system: traffic control. Front Cell Infect Microbiol 3, 4.

Dötsch, A., Eckweiler, D., Schniederjans, M., Zimmermann, A., Jensen, V., Scharfe, M., Geffers, R., and Häussler, S. (2012). The Pseudomonas aeruginosa transcriptome in planktonic cultures and static biofilms using RNA sequencing. PLoS ONE 7, e31092.

Duan, R., Liang, J., Shi, G., Cui, Z., Hai, R., Wang, P., Xiao, Y., Li, K., Qiu, H., Gu, W., et al. (2014). Homology analysis of pathogenic Yersinia species Yersinia enterocolitica, Yersinia

2 Introduction

pseudotuberculosis, and Yersinia pestis based on multilocus sequence typing. J. Clin. Microbiol. 52, 20–29.

Durand, E.A., Maldonado-Arocho, F.J., Castillo, C., Walsh, R.L., and Mecsas, J. (2010). The presence of professional phagocytes dictates the number of host cells targeted for Yop translocation during infection. Cell. Microbiol. 12, 1064–1082.

European Food Safety Authority (EFSA) (2015). The European Union summary report on trends and sources of zoonoses , trends and sources of zoonoses , zoonotic agents and food-borne outbreaks in 2015.

Eitel, J., and Dersch, P. (2002). The YadA protein of Yersinia pseudotuberculosis mediates high- efficiency uptake into human cells under environmental conditions in which invasin is repressed. Infect. Immun. 70, 4880–4891.

Erridge, C., Bennett-Guerrero, E., and Poxton, I.R. (2002). Structure and function of lipopolysaccharides. Microbes Infect. 4, 837–851.

Falcão, J.P., Falcão, D.P., Pitondo-Silva, A., Malaspina, A.C., and Brocchi, M. (2006). Molecular typing and virulence markers of Yersinia enterocolitica strains from human, animal and food origins isolated between 1968 and 2000 in Brazil. J. Med. Microbiol. 55, 1539–1548.

Foley, J.A., and Mathews, J.A. (1984). Reactive arthritis due to Yersinia enterocolitica. Clin. Rheumatol. 3, 385–387.

Fonseca, D.M. da, Hand, T.W., Han, S.-J., Gerner, M.Y., Glatman Zaretsky, A., Byrd, A.L., Harrison, O.J., Ortiz, A.M., Quinones, M., Trinchieri, G., et al. (2015). Microbiota-Dependent Sequelae of Acute Infection Compromise Tissue-Specific Immunity. Cell 163, 354–366.

Forman, S., Paulley, J.T., Fetherston, J.D., Cheng, Y.-Q., and Perry, R.D. (2010). Yersinia ironomics: comparison of iron transporters among Yersinia pestis biotypes and its nearest neighbor, Yersinia pseudotuberculosis. Biometals 23, 275–294.

Foultier, B., Troisfontaines, P., Müller, S., Opperdoes, F.R., and Cornelis, G.R. (2002). Characterization of the ysa pathogenicity locus in the chromosome of Yersinia enterocolitica and phylogeny analysis of type III secretion systems. J. Mol. Evol. 55, 37–51.

Foultier, B., Troisfontaines, P., Vertommen, D., Marenne, M.-N., Rider, M., Parsot, C., and Cornelis, G.R. (2003). Identification of substrates and chaperone from the Yersinia enterocolitica 1B Ysa type III secretion system. Infect. Immun. 71, 242–253.

Fredriksson-Ahomaa, M., Stolle, A., Siitonen, A., and Korkeala, H. (2006). Sporadic human Yersinia enterocolitica infections caused by bioserotype 4/O : 3 originate mainly from pigs. J. Med. Microbiol. 55, 747–749.

Fröhlich, K.S., Papenfort, K., Berger, A.A., and Vogel, J. (2012). A conserved RpoS-dependent small RNA controls the synthesis of major porin OmpD. Nucleic Acids Res. 40, 3623–3640.

2 Introduction

Fukushima, H., Gomyoda, M., Shiozawa, K., Kaneko, S., and Tsubokura, M. (1988). Yersinia pseudotuberculosis infection contracted through water contaminated by a wild animal. J. Clin. Microbiol. 26, 584–585.

Georg, J., and Hess, W.R. (2011). cis-antisense RNA, another level of gene regulation in bacteria. Microbiol. Mol. Biol. Rev. 75, 286–300.

Giangrossi, M., Prosseda, G., Tran, C.N., Brandi, A., Colonna, B., and Falconi, M. (2010). A novel antisense RNA regulates at transcriptional level the virulence gene icsA of Shigella flexneri. Nucleic Acids Res. 38, 3362–3375.

Gómez-Lozano, M., Marvig, R.L., Molin, S., and Long, K.S. (2012). Genome-wide identification of novel small RNAs in Pseudomonas aeruginosa. Environ. Microbiol. 14, 2006–2016.

Gong, H., Vu, G.-P., Bai, Y., Chan, E., Wu, R., Yang, E., Liu, F., and Lu, S. (2011). A Salmonella small non-coding RNA facilitates bacterial invasion and intracellular replication by modulating the expression of virulence factors. PLoS Pathog. 7, e1002120.

Gottesman, S., McCullen, C.A., Guillier, M., Vanderpool, C.K., Majdalani, N., Benhammou, J., Thompson, K.M., FitzGerald, P.C., Sowa, N.A., and FitzGerald, D.J. (2006). Small RNA regulators and the bacterial response to stress. Cold Spring Harb. Symp. Quant. Biol. 71, 1–11.

Grützkau, A., Hanski, C., Hahn, H., and Riecken, E.O. (1990). Involvement of M cells in the bacterial invasion of Peyer’s patches: a common mechanism shared by Yersinia enterocolitica and other enteroinvasive bacteria. Gut 31, 1011–1015.

Håkansson, S., Bergman, T., Vanooteghem, J.C., Cornelis, G., and Wolf-Watz, H. (1993). YopB and YopD constitute a novel class of Yersinia Yop proteins. Infect. Immun. 61, 71–80.

Haller, J.C., Carlson, S., Pederson, K.J., and Pierson, D.E. (2000). A chromosomally encoded type III secretion pathway in Yersinia enterocolitica is important in virulence. Mol. Microbiol. 36, 1436–1446.

Hamburger, Z.A., Brown, M.S., Isberg, R.R., and Bjorkman, P.J. (1999). Crystal structure of invasin: a bacterial integrin-binding protein. Science 286, 291–295.

Hammer, B.K., and Bassler, B.L. (2007). Regulatory small RNAs circumvent the conventional quorum sensing pathway in pandemic Vibrio cholerae. Proc Natl. Acad. Sci. U.S.A. 104,11145–11149.

Hansen-Wester, I., and Hensel, M. (2001). Salmonella pathogenicity islands encoding type III secretion systems. Microbes Infect. 3, 549–559.

Hartung, M., and Gerigk, K. (1991). Yersinia in effluents from the food-processing industry. Rev. - Off. Int. Epizoot. 10, 799–811.

Heesemann, J., Hantke, K., Vocke, T., Saken, E., Rakin, A., Stojiljkovic, I., and Berner, R. (1993). Virulence of Yersinia enterocolitica is closely associated with siderophore production, expression of an iron-repressible outer membrane polypeptide of 65,000 Da and pesticin sensitivity. Mol. Microbiol. 8, 397–408.

2 Introduction

Heesemann, J., Sing, A., and Trülzsch, K. (2006). Yersinia’s stratagem: targeting innate and adaptive immune defense. Curr. Opin. Microbiol. 9, 55–61.

Heim, F., Fehlhaber, K., and Scheibner, G. (1984). [The behavior of Yersinia enterocolitica at different temperatures and at various concentrations of curing salt].Arch ExpVeterinarmed38,729–734.

Heine, W., Beckstette M., Heroven A.K., Thiemann S., Heise U., Nuss A.M., Pisano F., Strowig T., Dersch P. (2018). Loss of CNFY toxin-induced inflammation drives Yersinia pseudotuberculosis into persistency. PloS Pathog. 14, e1006858.

Heise, T., and Dersch, P. (2006). Identification of a domain in Yersinia virulence factor YadA that is crucial for extracellular matrix-specific cell adhesion and uptake. Proc. Natl. Acad. Sci. U.S.A. 103, 3375–3380. al-Hendy, A., Toivanen, P., and Skurnik, M. (1992). Lipopolysaccharide O side chain of Yersinia enterocolitica O:3 is an essential virulence factor in an orally infected murine model. Infect. Immun. 60, 870–875.

Herbst, K., Bujara, M., Heroven, A.K., Opitz, W., Weichert, M., Zimmermann, A., and Dersch, P. (2009). Intrinsic thermal sensing controls proteolysis of Yersinia virulence regulator RovA. PLoS Pathog. 5, e1000435.

Heroven, A.K., Nagel, G., Tran, H.J., Parr, S., and Dersch, P. (2004). RovA is autoregulated and antagonizes H-NS-mediated silencing of invasin and rovA expression in Yersinia pseudotuberculosis. Mol. Microbiol. 53, 871–888.

Hoiczyk, E., Roggenkamp, A., Reichenbecher, M., Lupas, A., and Heesemann, J. (2000). Structure and sequence analysis of Yersinia YadA and Moraxella UspAs reveal a novel class of adhesins. EMBO J. 19, 5989–5999.

Hu, Z., Patel, I.R., and Mukherjee, A. (2013). Genetic analysis of the roles of agaA, agaI, and agaS genes in the N-acetyl-D-galactosamine and D-galactosamine catabolic pathways in Escherichia coli strains O157:H7 and C. BMC Microbiol. 13, 94.

Hurst, M.R.H., Becher, S.A., Young, S.D., Nelson, T.L., and Glare, T.R. (2011). Yersinia entomophaga sp. nov., isolated from the New Zealand grass grub Costelytra zealandica. Int. J. Syst. Evol. Microbiol. 61, 844–849.

Irnov, I., Sharma, C.M., Vogel, J., and Winkler, W.C. (2010). Identification of regulatory RNAs in Bacillus subtilis. Nucleic Acids Res. 38, 6637–6651.

Isberg, R.R., Voorhis, D.L., and Falkow, S. (1987). Identification of invasin: a protein that allows enteric bacteria to penetrate cultured mammalian cells. Cell 50, 769–778.

Jalava, K., Hakkinen, M., Valkonen, M., Nakari, U.-M., Palo, T., Hallanvuo, S., Ollgren, J., Siitonen, A., and Nuorti, J.P. (2006). An outbreak of gastrointestinal illness and erythema nodosum from grated carrots contaminated with Yersinia pseudotuberculosis. J. Infect. Dis. 194, 1209–1216.

Jones, K.E., Patel, N.G., Levy, M.A., Storeygard, A., Balk, D., Gittleman, J.L., and Daszak, P. (2008). Global trends in emerging infectious diseases. Nature 451, 990–993.

2 Introduction

Joutsen, S., Laukkanen-Ninios, R., Henttonen, H., Niemimaa, J., Voutilainen, L., Kallio, E.R., Helle, H., Korkeala, H., and Fredriksson-Ahomaa, M. (2017). Yersinia spp. in Wild Rodents and Shrews in Finland. Vector Borne Zoonotic Dis. 17, 303–311.

Kawano, M., Aravind, L., and Storz, G. (2007). An antisense RNA controls synthesis of an SOS- induced toxin evolved from an antitoxin. Mol. Microbiol. 64, 738–754.

Kenyon, J.J., Duda, K.A., De Felice, A., Cunneen, M.M., Molinaro, A., Laitinen, J., Skurnik, M., Holst, O., Reeves, P.R., and De Castro, C. (2016). Serotype O:8 isolates in the Yersinia pseudotuberculosis complex have different O-antigen gene clusters and produce various forms of rough LPS. Innate Immun 22, 205–217.

Keto-Timonen, R., Hietala, N., Palonen, E., Hakakorpi, A., Lindström, M., and Korkeala, H. (2016). Cold Shock Proteins: A Minireview with Special Emphasis on Csp-family of Enteropathogenic Yersinia. Front Microbiol 7, 1151.

Kim, T., Bak, G., Lee, J., and Kim, K.-S. (2015). Systematic analysis of the role of bacterial Hfq- interacting sRNAs in the response to antibiotics. J. Antimicrob. Chemother. 70, 1659–1668.

Kim, T.-J., Young, B.M., and Young, G.M. (2008). Effect of flagellar mutations on Yersinia enterocolitica biofilm formation. Appl. Environ. Microbiol. 74, 5466–5474.

Kolodziejek, A.M., Schnider, D.R., Rohde, H.N., Wojtowicz, A.J., Bohach, G.A., Minnich, S.A., and Hovde, C.J. (2010). Outer membrane protein X (Ail) contributes to Yersinia pestis virulence in pneumonic plague and its activity is dependent on the lipopolysaccharide core length. Infect. Immun. 78, 5233–5243.

Koo, J.T., Alleyne, T.M., Schiano, C.A., Jafari, N., and Lathem, W.W. (2011). Global discovery of small RNAs in Yersinia pseudotuberculosis identifies Yersinia-specific small, noncoding RNAs required for virulence. Proc. Natl. Acad. Sci. U.S.A. 108, E709-717.

Koornhof, H.J., Smego, R.A., and Nicol, M. (1999). Yersiniosis. II: The pathogenesis of Yersinia infections. Eur. J. Clin. Microbiol. Infect. Dis. 18, 87–112.

Kraehenbuhl, J.P., and Neutra, M.R. (2000). Epithelial M cells: differentiation and function. Annu. Rev. Cell Dev. Biol. 16, 301–332.

Laitenen, O., Tuuhea, J., and Ahvonen, P. (1972). Polyarthritis associated with Yersinia enterocolitica infection. Clinical features and laboratory findings in nine cases with severe joint symptoms. Ann. Rheum. Dis. 31, 34–39.

Laitinen, O., Leirisalo, M., and Skylv, G. (1977). Relation between HLA-B27 and clinical features in patients with yersinia arthritis. Arthritis Rheum. 20, 1121–1124.

Le Guern, A.-S., Martin, L., Savin, C., and Carniel, E. (2016). Yersiniosis in France: overview and potential sources of infection. Int. J. Infect. Dis. 46, 1–7.

2 Introduction

Leon-Velarde, C.G., Happonen, L., Pajunen, M., Leskinen, K., Kropinski, A.M., Mattinen, L., Rajtor, M., Zur, J., Smith, D., Chen, S., et al. (2016). Yersinia enterocolitica-Specific Infection by Bacteriophages TG1 and ϕR1-RT Is Dependent on Temperature-Regulated Expression of the Phage Host Receptor OmpF. Appl. Environ. Microbiol. 82, 5340–5353.

Liu, M., Alice, A.F., Naka, H., and Crosa, J.H. (2007). The HlyU protein is a positive regulator of rtxA1, a gene responsible for cytotoxicity and virulence in the human pathogen Vibrio vulnificus. Infect. Immun. 75, 3282–3289.

Liu, M.Y., Gui, G., Wei, B., Preston, J.F., Oakford, L., Yüksel, U., Giedroc, D.P., and Romeo, T. (1997). The RNA molecule CsrB binds to the global regulatory protein CsrA and antagonizes its activity in Escherichia coli. J. Biol. Chem. 272, 17502–17510.

MacDonald, E., Einöder-Moreno, M., Borgen, K., Thorstensen Brandal, L., Diab, L., Fossli, Ø., Guzman Herrador, B., Hassan, A.A., Johannessen, G.S., Johansen, E.J., et al. (2016). National outbreak of Yersinia enterocolitica infections in military and civilian populations associated with consumption of mixed salad, Norway, 2014. Euro Surveill. 21.

Macnab, R.M. (1999). The bacterial flagellum: reversible rotary propellor and type III export apparatus. J. Bacteriol. 181, 7149–7153.

Magistrali, C.F., Cucco, L., Manuali, E., Sebastiani, C., Farneti, S., Ercoli, L., and Pezzotti, G. (2014). Atypical Yersinia pseudotuberculosis serotype O:3 isolated from hunted wild boars in Italy. Vet. Microbiol. 171, 227–231.

Mäki-Ikola, O., Heesemann, J., Toivanen, A., and Granfors, K. (1997). High frequency of Yersinia antibodies in healthy populations in Finland and Germany. Rheumatol. Int. 16, 227–229.

Mandlik, A., Livny, J., Robins, W.P., Ritchie, J.M., Mekalanos, J.J., and Waldor, M.K. (2011). RNA-Seq-based monitoring of infection-linked changes in Vibrio cholerae gene expression. Cell Host Microbe 10, 165–174.

Marcus, S.L., Brumell, J.H., Pfeifer, C.G., and Finlay, B.B. (2000). Salmonella pathogenicity islands: big virulence in small packages. Microbes Infect. 2, 145–156.

Massé, E., Escorcia, F.E., and Gottesman, S. (2003). Coupled degradation of a small regulatory RNA and its mRNA targets in Escherichia coli. Genes Dev. 17, 2374–2383.

Matsumoto, H., and Young, G.M. (2006). Proteomic and functional analysis of the suite of Ysp proteins exported by the Ysa type III secretion system of Yersinia enterocolitica Biovar 1B. Mol. Microbiol. 59, 689–706.

Matsumoto, H., and Young, G.M. (2009). Translocated effectors of Yersinia. Curr. Opin. Microbiol. 12, 94–100.

Matsuura, M., Takahashi, H., Watanabe, H., Saito, S., and Kawahara, K. (2010). Immunomodulatory effects of Yersinia pestis lipopolysaccharides on human macrophages. Clin. Vaccine Immunol. 17, 49–55.

2 Introduction

McNally, A., Thomson, N.R., Reuter, S., and Wren, B.W. (2016). “Add, stir and reduce”: Yersinia spp. as model bacteria for pathogen evolution. Nat. Rev. Microbiol. 14, 177–190.

Mikula, K.M., Kolodziejczyk, R., and Goldman, A. (2012). Yersinia infection tools-characterization of structure and function of adhesins. Front Cell Infect Microbiol 2, 169.

Miller, E.F., and Maier, R.J. (2014). Ammonium metabolism enzymes aid Helicobacter pylori acid resistance. J. Bacteriol. 196, 3074–3081.

Miller, V.L., and Falkow, S. (1988). Evidence for two genetic loci in Yersinia enterocolitica that can promote invasion of epithelial cells. Infect. Immun. 56, 1242–1248.

Miller, V.L., Beer, K.B., Heusipp, G., Young, B.M., and Wachtel, M.R. (2001). Identification of regions of Ail required for the invasion and serum resistance phenotypes.Mol.Microbiol.41,1053–1062.

Moreau, K., Lacas-Gervais, S., Fujita, N., Sebbane, F., Yoshimori, T., Simonet, M., and Lafont, F. (2010). Autophagosomes can support Yersinia pseudotuberculosis replication in macrophages. Cell. Microbiol. 12, 1108–1123.

Mueller, C.A., Broz, P., Müller, S.A., Ringler, P., Erne-Brand, F., Sorg, I., Kuhn, M., Engel, A., and Cornelis, G.R. (2005). The V-antigen of Yersinia forms a distinct structure at the tip of injectisome needles. Science 310, 674–676.

Murros-Kontiainen, A., Fredriksson-Ahomaa, M., Korkeala, H., Johansson, P., Rahkila, R., and Björkroth, J. (2011a). Yersinia nurmii sp. nov. Int. J. Syst. Evol. Microbiol. 61, 2368–2372.

Murros-Kontiainen, A., Johansson, P., Niskanen, T., Fredriksson-Ahomaa, M., Korkeala, H., and Björkroth, J. (2011b). Yersinia pekkanenii sp. nov. Int. J. Syst. Evol. Microbiol. 61, 2363–2367.

Muszyński, A., Rabsztyn, K., Knapska, K., Duda, K.A., Duda-Grychtoł, K., Kasperkiewicz, K., Radziejewska-Lebrecht, J., Holst, O., and Skurnik, M. (2013). Enterobacterial common antigen and O-specific polysaccharide coexist in the lipopolysaccharide of Yersinia enterocolitica serotype O:3. Microbiology (Reading, Engl.) 159, 1782–1793.

Nagel, G., Lahrz, A., and Dersch, P. (2001). Environmental control of invasin expression in Yersinia pseudotuberculosis is mediated by regulation of RovA, a transcriptional activator of the SlyA/Hor family. Mol. Microbiol. 41, 1249–1269.

Najdenski, H., Golkocheva-Markova, E., Kussovski, V., Vesselinova, A., Garbom, S., and Wolf- Watz, H. (2009). Attenuation and preserved immunogenic potential of Yersinia pseudotuberculosis mutant strains evidenced in oral pig model. Zoonoses Public Health 56, 157–168.

Nakamura, S., Hayashidani, H., Iwata, T., Takada, M., and Une, Y. (2009). Spontaneous Yersiniosis due to Yersinia pseudotuberculosis serotype 7 in a squirrel monkey. J. Vet. Med. Sci. 71, 1657–1659.

Nakamura, S., Settai, S., Hayashidani, H., Urabe, T., Namai, S., and Une, Y. (2013). Outbreak of yersiniosis in Egyptian rousette bats (Rousettus aegyptiacus) caused by Yersinia pseudotuberculosis serotype 4b. J. Comp. Pathol. 148, 410–413.

2 Introduction

Narberhaus, F., Waldminghaus, T., and Chowdhury, S. (2006). RNA thermometers. FEMS Microbiol. Rev. 30, 3–16.

Neubauer, H., Aleksic, S., Hensel, A., Finke, E.J., and Meyer, H. (2000). Yersinia enterocolitica 16S rRNA gene types belong to the same genospecies but form three homology groups. Int. J. Med. Microbiol. 290, 61–64.

Nielsen, B., Heisel, C., and Wingstrand, A. (1996). Time course of the serological response to Yersinia enterocolitica O:3 in experimentally infected pigs. Vet. Microbiol. 48, 293–303.

Nitzan, M., Rehani, R., and Margalit, H. (2017). Integration of Bacterial Small RNAs in Regulatory Networks. Annu Rev Biophys 46, 131–148.

Nuorti, J.P., Niskanen, T., Hallanvuo, S., Mikkola, J., Kela, E., Hatakka, M., Fredriksson-Ahomaa, M., Lyytikainen, O., Siitonen, A., Korkeala, H., et al. (2004). A widespread outbreak of Yersinia pseudotuberculosis O:3 infection from iceberg lettuce. J. Infect. Dis. 189, 766–774.

Nuss, A.M., Heroven, A.K., Waldmann, B., Reinkensmeier, J., Jarek, M., Beckstette, M., and Dersch, P. (2015). Transcriptomic profiling of Yersinia pseudotuberculosis reveals reprogramming of the Crp regulon by temperature and uncovers Crp as a master regulator of small RNAs. PLoS Genet. 11, e1005087.

Nuss, A.M., Heroven, A.K., and Dersch, P. (2017). RNA Regulators: Formidable Modulators of Yersinia Virulence. Trends Microbiol. 25, 19–34.

Obana, N., Shirahama, Y., Abe, K., and Nakamura, K. (2010). Stabilization of Clostridium perfringens collagenase mRNA by VR-RNA-dependent cleavage in 5’ leader sequence. Mol. Microbiol. 77, 1416–1428.

Oliva, G., Sahr, T., and Buchrieser, C. (2015). Small RNAs, 5’ UTR elements and RNA-binding proteins in intracellular bacteria: impact on metabolism and virulence.FEMSMicrobiol.Rev.39,331–349.

Opdyke, J.A., Kang, J.-G., and Storz, G. (2004). GadY, a small-RNA regulator of acid response genes in Escherichia coli. J. Bacteriol. 186, 6698–6705.

O’Sullivan, T., Friendship, R., Blackwell, T., Pearl, D., McEwen, B., Carman, S., Slavić, D., and Dewey, C. (2011). Microbiological identification and analysis of swine tonsils collected from carcasses at slaughter. Can. J. Vet. Res. 75, 106–111.

Papenfort, K., and Vanderpool, C.K. (2015). Target activation by regulatory RNAs in bacteria. FEMS Microbiol. Rev. 39, 362–378.

Pappesch, R., Warnke, P., Mikkat, S., Normann, J., Wisniewska-Kucper, A., Huschka, F., Wittmann, M., Khani, A., Schwengers, O., Oehmcke-Hecht, S., et al. (2017). The Regulatory Small RNA MarS Supports Virulence of Streptococcus pyogenes. Sci Rep 7, 12241.

Park, J., Kim, S.M., Jeong, H.G., and Choi, S.H. (2012). Regulatory characteristics of the Vibrio vulnificus rtxHCA operon encoding a MARTX toxin. J. Microbiol. 50, 878–881.

2 Introduction

Pärn, T., Hallanvuo, S., Salmenlinna, S., Pihlajasaari, A., Heikkinen, S., Telkki-Nykänen, H., Hakkinen, M., Ollgren, J., Huusko, S., and Rimhanen-Finne, R. (2015). Outbreak of Yersinia pseudotuberculosis O:1 infection associated with raw milk consumption, Finland, spring 2014. Euro Surveill. 20.

Pelludat, C., Rakin, A., Jacobi, C.A., Schubert, S., and Heesemann, J. (1998). The yersiniabactin biosynthetic gene cluster of Yersinia enterocolitica: organization and siderophore-dependent regulation. J. Bacteriol. 180, 538–546.

Pepe, J.C., and Miller, V.L. (1993). Yersinia enterocolitica invasin: a primary role in the initiation of infection. Proc. Natl. Acad. Sci. U.S.A. 90, 6473–6477.

Pepe, J.C., Badger, J.L., and Miller, V.L. (1994). Growth phase and low pH affect the thermal regulation of the Yersinia enterocolitica inv gene. Mol. Microbiol. 11, 123–135.

Pepe, J.C., Wachtel, M.R., Wagar, E., and Miller, V.L. (1995). Pathogenesis of defined invasion mutants of Yersinia enterocolitica in a BALB/c mouse model of infection. Infect. Immun.63,4837–4848.

Perry, R.D., and Fetherston, J.D. (1997). Yersinia pestis--etiologic agent of plague. Clin. Microbiol. Rev. 10, 35–66.

Pettersson, J., Nordfelth, R., Dubinina, E., Bergman, T., Gustafsson, M., Magnusson, K.E., and Wolf-Watz, H. (1996). Modulation of virulence factor expression by pathogen target cell contact. Science 273, 1231–1233.

Pierson, D.E., and Falkow, S. (1990). Nonpathogenic isolates of Yersinia enterocolitica do not contain functional inv-homologous sequences. Infect. Immun. 58, 1059–1064.

Pierson, D.E., and Falkow, S. (1993). The ail gene of Yersinia enterocolitica has a role in the ability of the organism to survive serum killing. Infect. Immun. 61, 1846–1852.

Pinta, E., Li, Z., Batzilla, J., Pajunen, M., Kasanen, T., Rabsztyn, K., Rakin, A., and Skurnik, M. (2012). Identification of three oligo-/polysaccharide-specific ligases in Yersinia enterocolitica. Mol. Microbiol. 83, 125–136.

Portnoy, D.A., and Falkow, S. (1981). Virulence-associated plasmids from Yersinia enterocolitica and Yersinia pestis. J. Bacteriol. 148, 877–883.

Prentice, M.B., and Rahalison, L. (2007). Plague. Lancet 369, 1196–1207.

Prévost, K., Salvail, H., Desnoyers, G., Jacques, J.-F., Phaneuf, E., and Massé, E. (2007). The small RNA RyhB activates the translation of shiA mRNA encoding a permease of shikimate, a compound involved in siderophore synthesis. Mol. Microbiol. 64, 1260–1273.

Raghavan, R., Groisman, E.A., and Ochman, H. (2011). Genome-wide detection of novel regulatory RNAs in E. coli. Genome Res. 21, 1487–1497.

Rakin, A., Schneider, L., and Podladchikova, O. (2012). Hunger for iron: the alternative siderophore iron scavenging systems in highly virulent Yersinia. Front Cell Infect Microbiol 2, 151.

2 Introduction

Ray, W.K., and Larson, T.J. (2004). Application of AgaR repressor and dominant repressor variants for verification of a gene cluster involved in N-acetylgalactosamine metabolism in Escherichia coli K- 12. Mol. Microbiol. 51, 813–826.

Rebeil, R., Ernst, R.K., Gowen, B.B., Miller, S.I., and Hinnebusch, B.J. (2004). Variation in lipid A structure in the pathogenic yersiniae. Mol. Microbiol. 52, 1363–1373.

Reinés, M., Llobet, E., Llompart, C.M., Moranta, D., Pérez-Gutiérrez, C., and Bengoechea, J.A. (2012). Molecular basis of Yersinia enterocolitica temperature-dependent resistance to antimicrobial peptides. J. Bacteriol. 194, 3173–3188.

Reuter, S., Thomson, N.R., and McNally, A. (2012). Evolutionary dynamics of the Yersinia enterocolitica complex. Adv. Exp. Med. Biol. 954, 15–22.

Reuter, S., Connor, T.R., Barquist, L., Walker, D., Feltwell, T., Harris, S.R., Fookes, M., Hall, M.E., Petty, N.K., Fuchs, T.M., et al. (2014). Parallel independent evolution of pathogenicity within the genus Yersinia. Proc. Natl. Acad. Sci. U.S.A. 111, 6768–6773.

Reuter, S., Corander, J., de Been, M., Harris, S., Cheng, L., Hall, M., Thomson, N.R. and McNally A. (2015). Directional gene flow and ecological separation in Yersinia enterocolitica. Microb Genom 1, e000030.

Revell, P.A., and Miller, V.L. (2000). A chromosomally encoded regulator is required for expression of the Yersinia enterocolitica inv gene and for virulence. Mol. Microbiol. 35, 677–685.

Robert Koch Insitut, Epidemiologisches Bulletin, 13. Oktober 2006/Nr. 41

Robert Koch Insitut, Epidemiologisches Bulletin, 14. November 2003/Nr. 46

Robert Koch Insitut, Epidemiologisches Bulletin, 18. Mai 2015/Nr. 20

Robins-Browne, R.M., Tzipori, S., Gonis, G., Hayes, J., Withers, M., and Prpic, J.K. (1985). The pathogenesis of Yersinia enterocolitica infection in gnotobiotic piglets. J. Med. Microbiol. 19, 297–308.

Rosner, B.M., Stark, K., and Werber, D. (2010). Epidemiology of reported Yersinia enterocolitica infections in Germany, 2001-2008. BMC Public Health 10, 337.

Rosqvist, R., Magnusson, K.E., and Wolf-Watz, H. (1994). Target cell contact triggers expression and polarized transfer of Yersinia YopE cytotoxin into mammalian cells. EMBO J. 13, 964–972.

Ruckdeschel, K., Roggenkamp, A., Lafont, V., Mangeat, P., Heesemann, J., and Rouot, B. (1997). Interaction of Yersinia enterocolitica with macrophages leads to macrophage cell death through apoptosis. Infect. Immun. 65, 4813–4821.

Saberi, F., Kamali, M., Najafi, A., Yazdanparast, A., and Moghaddam, M.M. (2016). Natural antisense RNAs as mRNA regulatory elements in bacteria: a review on function and applications. Cell. Mol. Biol. Lett. 21, 6.

Salvail, H., and Massé, E. (2012). Regulating iron storage and metabolism with RNA: an overview of posttranscriptional controls of intracellular iron homeostasis. Wiley Interdiscip Rev RNA 3, 26–36.

2 Introduction

Sansonetti, P. (2002). Host-pathogen interactions: the seduction of molecular cross talk. Gut 50 Suppl 3, III2-8.

Sansonetti, P.J. (2004). War and peace at mucosal surfaces. Nat. Rev. Immunol. 4, 953–964.

Satchell, K.J.F. (2007). MARTX, multifunctional autoprocessing repeats-in-toxin toxins. Infect. Immun. 75, 5079–5084.

Satchell, K.J.F. (2011). Structure and function of MARTX toxins and other large repetitive RTX proteins. Annu. Rev. Microbiol. 65, 71–90.

Savin, C., Martin, L., Bouchier, C., Filali, S., Chenau, J., Zhou, Z., Becher, F., Fukushima, H., Thomson, N.R., Scholz, H.C., et al. (2014). The Yersinia pseudotuberculosis complex: characterization and delineation of a new species, Yersinia wautersii.Int.J.Med.Microbiol.304,452–463.

Schaake, J., Kronshage, M., Uliczka, F., Rohde, M., Knuuti, T., Strauch, E., Fruth, A., Wos-Oxley, M., and Dersch, P. (2013). Human and animal isolates of Yersinia enterocolitica show significant serotype-specific colonization and host-specific immune defense properties. Infect. Immun. 81, 4013– 4025.

Schaake, J., Drees, A., Grüning, P., Uliczka, F., Pisano, F., Thiermann, T., von Altrock, A., Seehusen, F., Valentin-Weigand, P., and Dersch, P. (2014). Essential role of invasin for colonization and persistence of Yersinia enterocolitica in its natural reservoir host, the pig. Infect. Immun. 82, 960–969.

Sedlyarova, N., Shamovsky, I., Bharati, B.K., Epshtein, V., Chen, J., Gottesman, S., Schroeder, R., and Nudler, E. (2016). sRNA-Mediated Control of Transcription Termination in E. coli. Cell 167, 111-121.e13.

Sharma, C.M., Darfeuille, F., Plantinga, T.H., and Vogel, J. (2007). A small RNA regulates multiple ABC transporter mRNAs by targeting C/A-rich elements inside and upstream of ribosome-binding sites. Genes Dev. 21, 2804–2817.

Sharma, C.M., Hoffmann, S., Darfeuille, F., Reignier, J., Findeiss, S., Sittka, A., Chabas, S., Reiche, K., Hackermüller, J., Reinhardt, R., et al. (2010). The primary transcriptome of the major human pathogen Helicobacter pylori. Nature 464, 250–255.

Shimoni, Y., Friedlander, G., Hetzroni, G., Niv, G., Altuvia, S., Biham, O., and Margalit, H. (2007). Regulation of gene expression by small non-coding RNAs: a quantitative view. Mol. Syst. Biol. 3, 138.

Sirisena, D.M., and Skurnik, M. (2003). Mutations in the genes for synthesis of the outer core region of the lipopolysaccharide of Yersinia enterocolitica O:3. J. Appl. Microbiol. 94, 686–692.

Skurnik, M., and Bengoechea, J.A. (2003). The biosynthesis and biological role of lipopolysaccharide O-antigens of pathogenic Yersiniae. Carbohydr. Res. 338, 2521–2529.

Skurnik, M., and Toivanen, P. (1993). Yersinia enterocolitica lipopolysaccharide: genetics and virulence. Trends Microbiol. 1, 148–152.

2 Introduction

Skurnik, M., Venho, R., Bengoechea, J.A., and Moriyón, I. (1999). The lipopolysaccharide outer core of Yersinia enterocolitica serotype O:3 is required for virulence and plays a role in outer membrane integrity. Mol. Microbiol. 31, 1443–1462.

Smego, R.A., Frean, J., and Koornhof, H.J. (1999). Yersiniosis I: microbiological and clinicoepidemiological aspects of plague and non-plague Yersinia infections. Eur. J. Clin. Microbiol. Infect. Dis. 18, 1–15.

Song, T., Mika, F., Lindmark, B., Liu, Z., Schild, S., Bishop, A., Zhu, J., Camilli, A., Johansson, J., Vogel, J., et al. (2008). A new Vibrio cholerae sRNA modulates colonization and affects release of outer membrane vesicles. Mol. Microbiol. 70, 100–111.

Sonnleitner, E., and Haas, D. (2011). Small RNAs as regulators of primary and secondary metabolism in Pseudomonas species. Appl. Microbiol. Biotechnol. 91, 63–79.

Spinner, J.L., Seo, K.S., O’Loughlin, J.L., Cundiff, J.A., Minnich, S.A., Bohach, G.A., and Kobayashi, S.D. (2010). Neutrophils are resistant to Yersinia YopJ/P-induced apoptosis and are protected from ROS-mediated cell death by the type III secretion system. PLoS ONE 5, e9279.

Steglich, C., Futschik, M.E., Lindell, D., Voss, B., Chisholm, S.W., and Hess, W.R. (2008). The challenge of regulation in a minimal photoautotroph: non-coding RNAs in Prochlorococcus. PLoS Genet. 4, e1000173.

Storz, G., Altuvia, S., and Wassarman, K.M. (2005). An abundance of RNA regulators. Annu. Rev. Biochem. 74, 199–217.

Storz, G., Vogel, J., and Wassarman, K.M. (2011). Regulation by small RNAs in bacteria: expanding frontiers. Mol. Cell 43, 880–891.

Straley, S.C., and Perry, R.D. (1995). Environmental modulation of gene expression and pathogenesis in Yersinia. Trends Microbiol. 3, 310–317.

Sulakvelidze, A. (2000). Yersiniae other than Y. enterocolitica, Y. pseudotuberculosis, and Y. pestis: the ignored species. Microbes Infect. 2, 497–513.

Swaminathan, B., Harmon, M.C., and Mehlman, I.J. (1982). Yersinia enterocolitica. J. Appl. Bacteriol. 52, 151–183.

Tacket, C.O., Narain, J.P., Sattin, R., Lofgren, J.P., Konigsberg, C., Rendtorff, R.C., Rausa, A., Davis, B.R., and Cohen, M.L. (1984). A multistate outbreak of infections caused by Yersinia enterocolitica transmitted by pasteurized milk. JAMA 251, 483–486.

Tennant, S.M., Grant, T.H., and Robins-Browne, R.M. (2003). Pathogenicity of Yersinia enterocolitica biotype 1A. FEMS Immunol. Med. Microbiol. 38, 127–137.

Thomson, N.R., Howard, S., Wren, B.W., Holden, M.T.G., Crossman, L., Challis, G.L., Churcher, C., Mungall, K., Brooks, K., Chillingworth, T., et al. (2006). The complete genome sequence and comparative genome analysis of the high pathogenicity Yersinia enterocolitica strain 8081. PLoS Genet. 2, e206.

2 Introduction

Tsai, C.-S., and Winans, S.C. (2011). The quorum-hindered transcription factor YenR of Yersinia enterocolitica inhibits pheromone production and promotes motility via a small non-coding RNA. Mol. Microbiol. 80, 556–571.

Tsolis, R.M., Young, G.M., Solnick, J.V., and Bäumler, A.J. (2008). From bench to bedside: stealth of enteroinvasive pathogens. Nat. Rev. Microbiol. 6, 883–892.

Tuompo, R., Hannu, T., Huovinen, E., Sihvonen, L., Siitonen, A., and Leirisalo-Repo, M. (2017). Yersinia enterocolitica biotype 1A: a possible new trigger of reactive arthritis. Rheumatol. Int. 37, 1863–1869.

Uliczka, F., Pisano, F., Schaake, J., Stolz, T., Rohde, M., Fruth, A., Strauch, E., Skurnik, M., Batzilla, J., Rakin, A., et al. (2011). Unique cell adhesion and invasion properties of Yersinia enterocolitica O:3, the most frequent cause of human Yersiniosis. PLoS Pathog. 7, e1002117.

Urban, J.H., and Vogel, J. (2008). Two seemingly homologous noncoding RNAs act hierarchically to activate glmS mRNA translation. PLoS Biol. 6, e64.

Valentin-Weigand, P., Heesemann, J., and Dersch, P. (2014). Unique virulence properties of Yersinia enterocolitica O:3--an emerging zoonotic pathogen using pigs as preferred reservoir host. Int. J. Med. Microbiol. 304, 824–834.

Vasala, M., Hallanvuo, S., Ruuska, P., Suokas, R., Siitonen, A., and Hakala, M. (2014). High frequency of reactive arthritis in adults after Yersinia pseudotuberculosis O:1 outbreak caused by contaminated grated carrots. Ann. Rheum. Dis. 73, 1793–1796.

Vecerek, B., Moll, I., and Bläsi, U. (2007). Control of Fur synthesis by the non-coding RNA RyhB and iron-responsive decoding. EMBO J. 26, 965–975.

Venecia, K., and Young, G.M. (2005). Environmental regulation and virulence attributes of the Ysa type III secretion system of Yersinia enterocolitica biovar 1B. Infect. Immun. 73, 5961–5977.

Viboud, G.I., and Bliska, J.B. (2005). Yersinia outer proteins: role in modulation of host cell signaling responses and pathogenesis. Annu. Rev. Microbiol. 59, 69–89.

Vockenhuber, M.-P., Sharma, C.M., Statt, M.G., Schmidt, D., Xu, Z., Dietrich, S., Liesegang, H., Mathews, D.H., and Suess, B. (2011). Deep sequencing-based identification of small non-coding RNAs in Streptomyces coelicolor. RNA Biol 8, 468–477.

Vogel, J. (2009). A rough guide to the non-coding RNA world of Salmonella. Mol. Microbiol. 71, 1–11.

Wacheck, S., Fredriksson-Ahomaa, M., König, M., Stolle, A., and Stephan, R. (2010). Wild boars as an important reservoir for foodborne pathogens. Foodborne Pathog. Dis. 7, 307–312.

Wachtel, M.R., and Miller, V.L. (1995). In vitro and in vivo characterization of an ail mutant of Yersinia enterocolitica. Infect. Immun. 63, 2541–2548.

Wagner, S., Sorg, I., Degiacomi, M., Journet, L., Dal Peraro, M., and Cornelis, G.R. (2009). The helical content of the YscP molecular ruler determines the length of the Yersinia injectisome. Mol. Microbiol. 71, 692–701.

2 Introduction

Waters, L.S., and Storz, G. (2009). Regulatory RNAs in bacteria. Cell 136, 615–628.

Weilbacher, T., Suzuki, K., Dubey, A.K., Wang, X., Gudapaty, S., Morozov, I., Baker, C.S., Georgellis, D., Babitzke, P., and Romeo, T. (2003). A novel sRNA component of the carbon storage regulatory system of Escherichia coli. Mol. Microbiol. 48, 657–670.

Westermark, L., Fahlgren, A., and Fällman, M. (2014). Yersinia pseudotuberculosis efficiently escapes polymorphonuclear neutrophils during early infection. Infect. Immun. 82, 1181–1191.

Williamson, D.A., Baines, S.L., Carter, G.P., da Silva, A.G., Ren, X., Sherwood, J., Dufour, M., Schultz, M.B., French, N.P., Seemann, T., et al. (2016). Genomic Insights into a Sustained National Outbreak of Yersinia pseudotuberculosis. Genome Biol Evol 8, 3806–3814.

Wren, B.W. (2003). The yersiniae--a model genus to study the rapid evolution of bacterial pathogens. Nat. Rev. Microbiol. 1, 55–64.

Wuthe, H.H., and Aleksić, S. (1997). [Yersinia enterocolitica serovar 2a, wb, 3:b,c biovar 5 in hares and sheep]. Berl. Munch. Tierarztl. Wochenschr. 110, 176–177.

Xia, L., Xia, W., Li, S., Li, W., Liu, J., Ding, H., Li, J., Li, H., Chen, Y., Su, X., et al. (2012). Identification and expression of small non-coding RNA, L10-Leader, in different growth phases of Streptococcus mutans. Nucleic Acid Ther 22, 177–186.

Yamashita, S., Lukacik, P., Barnard, T.J., Noinaj, N., Felek, S., Tsang, T.M., Krukonis, E.S., Hinnebusch, B.J., and Buchanan, S.K. (2011). Structural insights into Ail-mediated adhesion in Yersinia pestis. Structure 19, 1672–1682.

Yan, Y., Su, S., Meng, X., Ji, X., Qu, Y., Liu, Z., Wang, X., Cui, Y., Deng, Z., Zhou, D., et al. (2013). Determination of sRNA expressions by RNA-seq in Yersinia pestis grown in vitro and during infection. PLoS ONE 8, e74495.

Young, B.M., and Young, G.M. (2002). Evidence for targeting of Yop effectors by the chromosomally encoded Ysa type III secretion system of Yersinia enterocolitica. J. Bacteriol. 184, 5563–5571.

Young, G.M., Amid, D., and Miller, V.L. (1996). A bifunctional urease enhances survival of pathogenic Yersinia enterocolitica and Morganella morganii at low pH. J. Bacteriol. 178, 6487–6495.

Young, G.M., Badger, J.L., and Miller, V.L. (2000). Motility is required to initiate host cell invasion by Yersinia enterocolitica. Infect. Immun. 68, 4323–4326.

Zhang, Y., Kiyohara, H., Matsumoto, T., and Yamada, H. (1997). Fractionation and chemical properties of immunomodulating polysaccharides from roots of Dipsacus asperoides. Planta Med. 63, 393–399.

Zhou, D., and Yang, R. (2009). Molecular Darwinian evolution of virulence in Yersinia pestis. Infect. Immun. 77, 2242–2250.

Ziolo, K.J., Jeong, H.-G., Kwak, J.S., Yang, S., Lavker, R.M., and Satchell, K.J.F. (2014). Vibrio vulnificus biotype 3 multifunctional autoprocessing RTX toxin is an adenylate cyclase toxin essential for virulence in mice. Infect. Immun. 82, 2148–2157.

3 Publication 1

Tissue dual RNA-seq allows fast discovery of infection- specific functions and riboregulators shaping host– pathogen transcriptomes

Aaron M. Nuss, Michael Beckstette, Maria Pimenova, Carina Schmühl, Wiebke Opitz, Fabio Pisano, Ann Kathrin Heroven, and Petra Dersch

Department of Molecular Infection Biology, Helmholtz Centre for Infection Research, 38124 Braunschweig, Germany

The paper was published in: Proc Natl Acad Sci USA. 2017 Jan 31; 114(5): E791–E800 doi: 10.1073/pnas.1613405114

The extent of contribution from Carina Schmühl was:

Cloning of mutagenesis plasmids Creating mutant strains: Y. pseudotuberculosis IP32953 ΔsgrS, IP32953 ΔryhB1/2, IP32953 ΔglmY, IP32953 ΔfruBKA, IP32953 ΔadhE In vitro analysis of deletion mutants Investigating effects of mutation in the in vivo mouse model: preparation of experiments; handling of organs after the sacrifice of mice; analysis of the results

3 Publication 1

4 Publication 2

Comparative transcriptomic profiling of Yersinia enterocolitica O:3 and O:8 reveals major expression differences of fitness- and virulence-relevant genes

indicating ecological separation

Carina Schmühl1, Michael Beckstette1, Ann Kathrin Heroven1, Boyke Bunk2, Cathrin Spröer2, Alan McNally3, Jörg Overmann2,4,5, and Petra Dersch1,4,5

1Department of Molecular Infection Biology, Helmholtz Centre for Infection Research,

Braunschweig, Germany 2Leibniz Institute DSMZ - German Collection of Microorganisms and Cell Cultures, Braunschweig,

Germany 3Institute of Microbiology and Infection, University of Birmingham, Edgbaston, Birmingham, B15 2TT,

UK 4 Department of Microbiology, Technical University Braunschweig 5German Center for Infection Research, Partner-site Hannover-Braunschweig, Germany

The paper was submitted to: Nucleid Acid Research Manuscript ID NAR-02763-2018

The extent of contribution from Carina Schmühl was:

Performing the experiments Analysing the data

4 Publication 2

4.1 Abstract

Yersinia enterocolitica is a zoonotic pathogen and an important cause of bacterial gastrointestinal infections in humans. Large-scale population genomic analyses revealed genetic and phenotypic diversity of this bacterial species, but little is known about the differences of the transcriptome organization, sRNA repertoire and the transcriptional output. Here, we present the first comparative high-resolution transcriptome analysis of Y. enterocolitica strains representing the high-pathogenic phylogroup 2 (serotype O:8) and the low-pathogenic phylogroup 3 (serotype O:3) grown under four infection-relevant conditions. Our RNA-Seq approach revealed 1299 and 1076 transcriptional start sites, and identified strain-specific sRNAs that could contribute to differential regulation among the phylogroups. Comparative transcriptomics further uncovered major gene expression differences, in particular of the temperature-responsive regulon. Multiple virulence-relevant genes are differentially regulated between both strains, indicating an ecological separation of the phylogroups with certain niche-adapted properties. Strong upregulation of the enterotoxin gene ystA in combination with a constitutive high expression of the cell invasion factor InvA further indicates that the toxicity of recent outbreak O:3 strains has increased. Overall, our study provides new insights into the phylogroup 2 and 3 specific transcriptome organization, and reveals gene expression differences that could modulate the substantial phenotypic variation that exists between the lineages.

4. 2 Introduction

The enteric pathogen Yersinia enterocolitica is the most common gram-negative zoonotic pathogen that leads to human yersiniosis, a variety of gut-associated diseases ranging from enteritis, watery diarrhea, mesenteric lymphadenitis to postinfectious extraintestinal sequelae such as reactive arthritis (Bottone, 1997; Valentin-Weigand et al., 2014). Yersiniosis is among the most common bacterial enteric diseases in the industrialized countries with the highest burden of disease in children under 15 years of age (Rosner et al., 2010; Valentin- Weigand et al., 2014). The species Y. enterocolitica constitutes a very diverse group of around 70 serotypes of which only 11 are harmful to humans. Among the isolated pathogenic strains are the highly mouse-virulent 1B/O:8 strains (YeO:8), recently classified into the phylogroup 2 (Reuter et al., 2014, 2015). This bioserotype, in particular YeO:8 strain 8081v has been used to study the pathogenesis of Y. enterocolitica using mouse infection models. However, by far the most frequent cause of human yersiniosis in Europe and Japan (>90%)

4 Publication 2

is Y. enterocolitica bioserotype O:3 (YeO:3), phylogroup 3, which is also frequently found in pigs and pork products (Bottone, 1997; Jones et al., 2003; Reuter et al., 2015). It primarily originates from domestic pigs, in which they colonize the lymphoid tissue of the gut and oropharynx mainly as asymptomatic (Fredriksson-Ahomaa et al., 2006a; Gürtler et al., 2005). YeO:3 is less common in North America, but has replaced YeO:8 (the most prevalent serotype in the 90th) and is now the predominat serotype (Tauxe, 2002; Fredriksson- Ahomaa et al., 2006). The reasons for ist rising global relevance are largely unknown, but recent genomic comparisons and analyses of their host colonization properties and fitness determinants (Reuter et al., 2014; Uliczka et al., 2011a). The gene content and synteny of YeO:8 and YeO:3 strains are largely conserved. However, recent studies also demonstrated that there are considerable genomic differences (Batzilla et al., 2011; Reuter et al., 2014). YeO:3 does not contain certain pathogenicity factors of YeO:8, such as the high-pathogenicity island (HPI) involved in the yersiniabactin (Ybt)-mediated iron uptake, and the chromosomally encoded ysa type III secretion system (T3SS). Instead, it has evolved an alternative set of virulence-associated traits including a RtxA-like toxin, two putative invasion-associated genes, two clusters of putative b-fimbriae, a dual functional insecticidal toxin, and another distinct ysp T3SS. The ysp T3SS is homologous to the Salmonella SPI-2 T3SS, but lacks functional parts; i.e. there are no effector genes linked to the T3SS gene cluster (Batzilla et al., 2011). Moreover, the rtxA toxin gene of the Rtx cluster is intact, but the secretion genes are disrupted, leaving involvement of ysp and rtx genes in pathogenicity unclear. Moreover, the pYV plasmids are more divergent than the corresponding genome sequences, and the cluster for the lipopolysaccharide (LPS) outer core and the O-antigen are differently organized. The distinct virulence traits trigger a different cytokine profile by primary human, porcine and murine macrophages. YeO:3 promotes a significant lower production of IL-8, but a considerably higher secretion of IL-10 (Schaake et al., 2013). It is likely that this contributes to inhibiting inflammation, immunopathological changes and favoring long-term persistence without severe clinical manifestations. Genomic variations of YeO:3 that streamline the physiology and metabolism and increase the overall fitness of the bacteria to their life-style are also likely to contribute to their worldwide success. For instance, YeO:3, but not YeO:8 strains, possess the aga operon that allows them to grow on N-acetyl-galactosamine (GalNAc) (Batzilla et al., 2011; Rakin et al., 2012). As GalNAc is the major amino sugar of porcine mucin, this metabolic trait may represent an important virulence-relevant fitness factor reflecting the adaptation of YeO:3 to

4 Publication 2

its prefered reservoir host, the pig. In addition, several genomic islands as well as the spectrum and number of prophages and insertion (IS) elements are different between YeO:3 and YeO:8. A wide variety of IS families (IS3, IS4 and IS200) dominate in YeO:8, whereas numerous ISYen1/IS1667 and ISYen2 elements are found in YeO:3. It is assumed that the genetic diversity of both serotypes leads to serotype-specific colonization and host-specific immune defense properties with different clinical outcome (Schaake et al., 2013), but how the serotype-specific characteristics impact pathogenicity in their prefered hosts is largely unclear. A comparative study of human-, pig- and food-derived YeO:3 isolates with YeO:8 revealed that identical colonization factors participate in host cell binding and invasion, yet, small genetic variations lead to profound changes of their expression pattern (Schaake et al., 2014; Uliczka et al., 2011a). An insertion of an ISYen1/IS1667 element caused a high and constitutive expression of the primary cell binding and invasion factor InvA, and a base pair substitution results in the synthesis of a stable variant of the InvA regulator RovA in YeO:3. Both changes have a significant effect upon host cell invasion and virulence in mice and pigs (Schaake et al., 2014; Uliczka et al., 2011a). To gain insight into the dimension of the transcriptional variability that correlates with the genomic and phenotypic differences of the O:3 and O:8 serotype, we used a comparative RNA-seq-based transcriptomic approach to identify serotype/isolate-specific differences in the transcriptome under infection-relevant conditions. This strategy allowed us to obtain the first in-depth single-nucleotide resolution transcriptome of Y. enterocolitica (including genome-wide promoter maps and the non-coding RNA repertoire), enabled us to reveal major differences in the temperature- and growth phase-dependent expression profiles, and led to the discovery of changes that modulate transcript levels of important virulence-relevant traits.

4.3 Material and Methods

Bacterial strains

All Y. enterocolitica strains were grown in Luria Broth (LB) to exponential phase (OD600 0.5) or stationary phase (16 h) at 25°C and 37°C under anaerobic conditions for RNA isolation and RNA-Seq analysis. Bacteria were cultivated in BHI medium for transformation with indicated plasmids. E. coli was grown at 37°C in LB medium. If necessary, antibiotics were

4 Publication 2

added at the following concentrations: kanamycin 50 µg ml-1, chloramphenicol 30 µg ml-1. All strains used in this study are listed in Table S4.1.

DNA manipulation and plasmid construction PCR amplification, restriction digestions, ligations and transformations were performed using standard genetic and molecular techniques (Miller 1992, Sambrook 2001). The plasmids used in this work are listed in Table S4.1. Oligonucleotides used for PCR and qRT-PCR were purchased from Metabion and listed in Table S4.1. Plasmid DNA was isolated using Nucleospin® Plasmid Kit (Macherey & Nagel, Germany). Restriction enzymes and DNA- modifying enzymes were purchased from New England Biolabs. PCRs were performed in a 50 µl volume for 29 cycles using Phusion High-Fidelity DNA polymerase (New England Biolabs) or Taq polymerase (Promega). Purification of PCR products was performed using the Nucleospin® Gel and PCR Clean-up (Macherey & Nagel, Germany). The resulting plasmids were sequenced by Seqlab (Göttingen, Germany). Plasmids pCS71, pCS72 and pCS63 were constructed by amplifying the 5’-UTR of ystA from genomic DNA of YeO:3 Y1 with forward primers VIII009, VIII010, VIII011 and reverse primer VIII016. Plasmids pCS61, pCS68 and pCS70 were constructed by amplifying the 5’-UTR of ystA from genomic DNA of YeO:8 8081v with forward primers VIII009, VIII014, VIII015 and reverse primer VIII019. The PCR-derived fragments were integrated into the XhoI/NheI site of pFU55 (Uliczka et al., 2011b), creating fusions of the 5’-UTR to lacZ. Plasmid pHT109 was constructed by amplifying the rovA gene (with its own promotor) using primers 123 and 508. The fragment was inserted into pZA31 using the KpnI and ClaI restriction sites.

YeO:3 strain Y1 genome sequencing and annotation Y. enterocolitica strain Y1 deposited at the DSMZ no. 107832 (NCBI: CP030980; CP030981), a recent O:3/4 human isolate (Uliczka et al., 2011a), was selected as reference strain and sequenced. Genomic DNA of Y1 was isolated using the Qiagen genomic-tip 100/G Kit (Qiagen, Germany). DNA concentration was measured using the Qubit Fluorometric Quantitation System (Thermo Fischer Scientific, USA) and adequate quality was verified using pulse field gel electrophoresis. The genomic sequence was determined using PacBio RSII and Illumina Hiseq2500. SMRTbell™ template library was prepared according to the instructions from PacificBiosciences, Menlo Park, CA, USA, following the Procedure & Checklist – Greater Than 10 kb Template Preparation. Briefly, for preparation of 15 kb libraries 8 µg genomic

4 Publication 2

DNA was sheared using g-tubes™ from Covaris, Woburn, MA, USA according to the manufacturer´s instructions. DNA was end-repaired and ligated overnight to hairpin adapters applying components from the DNA/Polymerase Binding Kit P6 from Pacific Biosciences, Menlo Park, CA, USA. Reactions were carried out according to the manufacturer´s instructions. BluePippin™ Size-Selection to greater than 4 kb was performed according to the manufacturer´s instructions (Sage Science, Beverly, MA, USA). Conditions for annealing of sequencing primers and binding of polymerase to purified SMRTbell™ template were assessed with the Calculator in RS Remote, PacificBiosciences, Menlo Park, CA, USA. SMRT sequencing was carried out on the PacBio RSII (PacificBiosciences, Menlo Park, CA, USA) taking one 240-minutes movie. The PacBio run yielded 70,767 reads with a mean read length of 12,720 bp. SMRT Cell data was assembled using the “RS_HGAP_Assembly.3“ protocol included in SMRT Portal version 2.3.0 using default parameters. The assembly revealed a circular chromosome (YEY1_1) and one circular plasmid (YEY1_2). Both replicons were circularized, particularly artificial redundancies at the ends of the contigs were removed and adjusted to dnaA and sopB as first genes. Error-correction was performed by a mapping of 1.4 Mio paired-end reads of 2 x 301 bp generated on an Illumina MiSeq onto finished genomes using BWA (Li and Durbin, 2009) with subsequent variant and consensus calling using VarScan (Koboldt et al., 2012). A consensus concordance of QV60 could be confirmed for the genome. Finally, an annotation was carried out using Prokka 1.8 (Seemann, 2014). Hereby, an optional user-provided set of annotated proteins was used as the primary source of annotation containing the annotation information of all genes in Y. pseudotuberculosis YPIII. The used annotation file is given as supplementary File S1. The average GC content is 47%, similar to that of Y. enterocolitica strain 8081 (NC_008800; 47.27%). The complete Y1 genome sequence was deposited in NCBI under accession numbers CP030980 (chromosome YEY1_1) and CP030981 (plasmid YEY1_2).

RNA isolation Y. enterocolitica 8081v and Y1 were grown in LB medium to exponential phase (OD600 0.5) or stationary phase (16 h) at 25°C and 37°C, respectively. Total bacterial RNA was isolated by a hot phenol extraction protocol (Sambrook, 2001). Remaining DNA was digested using the TURBOTM DNase (Ambion), and RNA was purified with phenol:chlorophorm:isoamylalcohol. The quality was assessed using the Agilent RNA 6000

4 Publication 2

Nano Kit on the Agilent 2100 Bioanalyzer (Agilent Technologies). From 5 µg of total RNA the rRNA was depleted using RiboZero (Illumina).

Strand-specific library preparation and Illumina sequencing Strand-specific RNA-seq cDNA library preparation and barcode introduction was performed using the NEBNext Multiplex Small RNA Library Prep Set for Illumina (New England Biolabs). In brief, the rRNA-depleted RNA was fragmented by sonication to a median size of 200 nt. The fragments were 5’ phosphorylated and ligated to 3’- and 5’-RNA-adapter oligonucleotides. After reverse transcription, cDNA libraries were PCR amplified (15 cycles). Quality of the libraries was validated using Agilent 2100 Bioanalyzer (Agilent Technologies) following the manufacturer’s instructions. Single-end sequencing on the HiSeq2500 was performed with 2 nM library denatured with 0,1 N NaOH and diluted to a final concentration of 12 pM. Cluster generation on HiSeqSR Flow Cell v3 was generated at cBot using TruSeq SR Cluster Kit v3 - HS to create single molecule DNA templates followed by bridge amplification. Sequencing run was performed at HiSeq2500 using TruSeq SBS Kit v3 (50 cycle) to run 51 cycles and 7 cycles for the single-indexed read. The fluorescent images were processed to sequences and transformed to FastQ format using the Genome Analyzer Pipeline Analysis software 1.8.2 (Illumina). The sequence output was controlled for general quality features, sequencing adapter clipping and demultiplexing using the fastq-mcf and fastq-multx tool of ea-utils: Command-line tools for processing biological data (Aronesty, 2011).

Read mapping, bioinformatics and statistics Quality of the sequencing output was analyzed using FastQC (Babraham Bioinformatics). All sequenced libraries were mapped to the YeO:8 8081v genome (NC_008800.1) and pYVO:8 plasmid (NC_008799.1) or the YeO:3 Y1 genome (CP030980) and pYVO:3 plasmid (CP030981) using fast gapped-read alignment tool Bowtie2 (Langmead and Salzberg, 2012) with default parameters. After read mapping, SAMtools (Li et al., 2009) was employed to filter the resulting bam files for uniquely mapped reads (both strands). Reads were classified as uniquely mapped reads with a unique genomic location if and only if they could not be aligned to another location with a higher or same mapping quality. The resulting bam files constituted the basis for all downstream analyses and were used for visualization. For detailed mapping statistics, see Dataset S1. Obtained data were further processed as described previously (Nuss et al., 2015, 2017).

4 Publication 2

Detection of transcriptional start sites To detect transcriptional start sites libraries treated with 5’ polyphosphatase (+ Phos) were compared to libraries not treated with 5’ polyphosphatase (- Phos), which provides the background distribution of read starts. The - Phos libraries are depleted for cDNA derived from fragments containing the 5’ end of primary transcripts, while the corresponding + Phos libraries are unbiased. To verify transcriptional start sites additional libraries treated with TEX (Terminator Exonuclease) were compared to - TEX libraries. TEX treated libraries are enriched for primary transcripts as TEX digests RNAs with 5’ monophosphate but not 5’ triphosphate. In a first step sample libraries were normalized to million uniquely mapped reads and for every base the coverage and the number of reads starting at the respective position were calculated. Then, biological replicates were combined/merged by averaging coverage and read starts data. To detect transcriptional start sites for YeO:8 8081v and YeO:3 Y1 we applied TSSAR (Amman et al., 2014) on the RNA-seq data. All TSSs obtained from TSSAR were inspected manually and curated. In case, a sharp 5' flank cDNA read (≥ 10 reads) with continuous coverage to a downstream gene was manually detected, this position was added to the set of mRNA TSSs, although TSSAR failed to detect the respective TSS. TSSs were assigned to four different categories (Schlüter et al., 2013): If a TSS was located upstream of an annotated gene it was assigned as mTSS (TSS of messenger RNA). When a TSS matches the position of translation start codon, but in proximity of 10 nt to the translational start codon, the TSS was annotated as lmTSS (TSS of leaderless transcript). As asTSSs (antisense RNAs) identified were TSSs of cis-encoded antisense RNAs, which are oriented antisense to a protein-coding gene with no continuous coverage to the gene located downstream. If a TSS was located in an intergenic region with appropriate distance and no coverage to the next start codon, it was assigned as sRNA (trans-encoded RNA). Adjacent TSS with a distance less than 3 nt were clustered and the TSS with the highest number of read start counts was annotated as TSS. The newly identified TSSs were labeled to following conventions, x_TSS_CDS_n, where “x” indicates strain Y1 or 8081v. TSS that were assigned to protein coding genes were compared between YeO:8 8081v and YeO:3 Y1 TSS were considered as conserved between the two strains if they are assigned to the same gene and located at the same distance (+/- 5 nt) to the translational start codon.

4 Publication 2

Detection of conserved sequence motifs To investigate potential sequence conservation at the determined TSS, a sequence logo for the +1 to +3 position (with the TSS being position +1) of all 1299 TSS for 8081v and 1076 TSS for Y1 was generated using the WebLogo Software (Crooks et al., 2004). We performed de-novo motif discovery using the MEME software (Bailey and Elkan, 1994) to compute conserved sequence motifs in the -10 and -35 promotor region. Subsequences starting at position -15 and ending at position -3 (relative to the TSS) of all TSSs determined for each strain served as input for motif detection in the -10 region. For the -35 region, we used subsequences starting at position -45 and ending -25. We ran MEME in Zero or One occurence per Sequence (ZOOPS) mode and searched for motifs between length three and eight for the -10 region and between length three and five for the -35 region.

Identification of small regulatory RNAs (sRNAs) To identify expressed sRNAs, a global screen in all samples for unannotated trans-encoded sRNAs and cis-encoded antisense RNAs was performed as described previously (Nuss et al., 2015). In brief, transcripts were assembled from reads and classified. For sRNA classification, TSS data were included in the Y. enterocolitica strain Y1 and 8081v annotation. In a first step, transcripts seeds, which correspond to genomic regions of minimal length of 40 nt and a continuous coverage of at least 30 reads were considered as candidates for sRNAs. The resultant transcripts were extended on both ends until the coverage was lower than 3 reads. Finally, transcripts located in intergenic regions without overlapping UTRs were classified as trans-encoded sRNAs, while transcripts found on the strand opposite to a protein-coding gene were defined as cis-encoded antisense RNAs. All sRNA candidates were inspected manually and checked if they passed this last filter. The novel non-coding RNAs were labelled according to the common convention (Ysr(e)_n) with ongoing numbers (n). Identified sRNA candidates were compared between YeO:8 8081v and YeO:3 Y1 based on BlastN analysis and the genomic context. Conservation of sRNAs within other Yersinia species and γ-proteobacteria was determined by BlastN analysis. RNA sequences were used to scan Rfam (Kalvari et al., 2018a, 2018b) for related sequences.

Differential expression analysis Reads aligned to annotated genes were quantified with the htseq-count program (Anders et al., 2015; Love et al., 2014). To detect genes that were differentially expressed in 8081v and Y1, we employed DESeq2 (version 1.2.1) (Love et al., 2014). For DESeq2 parameterization

4 Publication 2

we used a beta prior and disabled the Cook Distance cut off filtering. All other parameters remained unchanged. HTSeq in union count mode was used to generate raw read counts required by DESeq2 as basis for differential expression analysis. In addition, RPKM (reads per kilobase max. transcript length per million mapped reads) values were computed for each library from the raw gene counts. The list of DESeq2 determined differentially expressed genes (DEGs) was filtered with a conservative absolute log2-fold change cut-off of at least 2 and a cut-off for a multiple testing corrected p-value of at most 0.05. To assess platform dynamic range and the accuracy of fold-change response, we used ERCC RNA Spike-In Controls (Thermo Fisher Scientific). Spike-in control sequences were added to the reference genome/annotation prior to read alignment and read counts for spike- in controls were determined along with normal gene counts with program htseq-count.

Cross species analysis To allow for a comparison of the transcriptomes of Y. enterocolitica strain Y1 and strain 8081v and construction of a correspondency table of locus tags, we computed a bijective mapping between all coding genes by reciprocal-best BLASTP (Altschul et al., 1997) hits with an E-value cutoff of 1.0E-6. By using this mapping table we were able to construct raw read count matrices containing corresponding counts from both of the two species and could use them for cross species DEG analysis with DESeq2. To construct the core proteome and to compare the expression profiles of more than two Yersiniae transcriptomes (see Fig. 4.1C) we clustered all protein coding genes on the basis results from an all-versus-all BLAST comparison. More precisely, we computed the core proteome of N Yersiniae strains by finding cliques of size N in the graph of reciprocal-best BLASTP hits across species boundaries where, each clique contains exactly one member of each of the N involved strains. The set of identified cliques allowed us to construct a correspondency table for core genes of more than two strains reflecting ortologous gene relationships as it is e.g. necessary for the principal component analysis of the expression profiles of several strains shown in Fig. 4.1C.

Quantitative real-time RT-PCR (qRT-PCR) qRT-PCR for the validation of RNA-sequencing results was performed on total RNA samples isolated from bacterial cultures grown at 25°C and 37°C to exponential and stationary phase. For the detection of ystA in Y. enterocolitica isolates, total RNA was isolated using the SV total RNA isolation Kit (Promega). 35 µg of RNA were treated with 4 Units of DNAse

4 Publication 2

(Ambion) in a 50 µl reaction. Afterwards the reaction was purified using phenol:chloroform:isoamylalcohol. Contamination assessment was performed with PCR and the Agilent 2100 Bioanalyzer (Agilent Technologies). The amount of RNA for each sample was determined using the NanoDrop One Spectrophotometer (Thermo Fischer Scientific). qRT-PCR was performed using the SensiFastNoRox Kit (Bioline) with 25 ng/µl of the RNA samples according to the manufacturers instructions. qRT-PCR was performed in a Rotor- Gene Q lightcycler (Qiagen). Primers used for analyzing relative gene expression purchased from Metabion and are listed in Table S4.1. The genes sopB (validation) and gyrB (ystA expression) were used for normalization. Data analysis was performed with the Rotor-Gene Q Series Software. Relative gene expression was calculated as described earlier (Pfaffl, 2001). Primer efficiencies were determined experimentally using serial dilutions of genomic Y. enterocolitica Y1 and 8081v DNA. Primer efficiencies are: ystA (YEY1_01327 / YE8081_01824): 2.02; sopB (YEY1_04214 / YE8081_04390): 2.21; gyrB (YEY1_00004 / YE8081_04289): 2.04; ureA (YEY1_00981 / YE8081_00974): 1.98; metR (YEY1_03883 / YE8081_00252): 2.00; smfA (YEY1_03315 / YE8081_00789): 2.03; fimA-6 (YEY1_03976 / YE8081_00164): 2.05; glnH (YEY1_02796 / YE8081_02909): 2.13; astC (YEY1_01889 / YE8081_02525): 2.03; leuO (YEY1_00693 / YE8081_00670): 1.94; Ysr212: 2.07; Ysr109: 2.17; Ysr021: 1.96; Ysr060: 1.99; Ysr143: 2.11.

Analysis of reporter gene expression The β-galactosidase activity assay of the lacZ fusion constructs was measured as described previously (Nagel et al., 2001). The activity was calculated as following: β-galactosidase -1 -1 -1 -1 activity = OD 420nm * 6,648 * OD 600nm * t (min) * Vol (ml) .

4.4 Results and Discussion

4.4.1 Comparative RNA-seq of Y. enterocolitica O:8 and O:3

In order to obtain high resolution transcription profiles and identify transcripts that were differentially expressed between serotype O:8 and O:3 strains, we built upon the transcriptome of YeO:8 strain 8081v and YeO:3 strain Y1. Strain 8081v is a well- characterized representative of YeO:8. It is a widely distributed highly virulent isolate, which

4 Publication 2

has played an pivotal role in the analysis of Yersinia infection. Many of its virulence factors have been characterized in detail and our knowledge of virulence-relevant gene regulation and networks was mainly derived from this strain of which a complete genome sequence is available (RefSeq Accession Nos: NC_008800.1 chromosome, NC_008791 virulence plasmid) (Thomson et al., 2006). YeO:3 strain Y1 is a recent isolate from an outbreak in Germany isolated from a patient stool. Its cell adhesion and invasion property, survival in macrophages, induction of immune responses and its virulence in mouse and pig models has been characterized and compared with strain 8081v (Schaake et al., 2013, 2014; Uliczka et al., 2011b). To allow precise mapping of the RNA sequencing reads the genome of strain Y1 was sequenced de novo. Data were assembled into two circular replicons, one of 4,522,295 bp for the genome (GenBank Accession No. CP030980) and the other of 72,411 bp for the pYVO:3 plasmid (GenBank Accession No. CP030981), which were annotated and used for transcriptome profiling. A sequence comparison with other available YeO:3 strains revealed an average nucleotide identity of 99.9% to YeO:3 1203 and 99.83% to YeO:3 Y11 (Fig. S4.1), making it a perfect representative strain of the highly clonal phylogroup 3. In order to obtain a comprehensive image of the primary transcriptome, we used rRNA- depleted total RNA of YeO:8 strain 8081v and YeO:3 strain Y1 grown to exponential or stationary phase at 25°C or 37°C resembling alterations in temperatures and nutrient limitations encountered in the initial or later/ongoing stages of infection (Fig. 4.1A). A global RNA-seq approach was employed by comparing mapped sequence reads from different strand-specific barcoded cDNA libraries of three independent biological replicates of five pooled cultures for each growth condition to catalog the transcripts for a detailed gene map. From each library between 727,853 and 3,015,896 uniquely mapped sequence reads were generated and mapped to the Y. enterocolitica 8081v genome (RefSeq Accession Nos.: chromosome NC_008791.1; virulence plasmid pYVO:8; NC_008800.1) or the Y. enterocolitica Y1 genome (RefSeq Accession Nos.: chromosome CP030980; virulence plasmid pYVO:3 CP030981) (Dataset S4.1). This is a sufficient coverage and robust representation of the Y. enterocolitica transcriptome in each of the four conditions (Fig. 4.1B, Dataset S4.1). The global gene expression profiles of both strains were distinct and the three biological triplicates clustered together (Fig. 4.1C).

4 Publication 2

A (3 pools/5 samples per pool) 25°C 37°C

exponential stationary exponential stationary

total RNA isolation rRNA depletion TAP-treatment RNA-seq cDNA library preparation Deep sequencing

B YeO:3 Y1 Y1_25_exp 8081_25_exp YeO:8 8081v pYV Y1_25_stat 8081_25_stat pYV Y1_37_exp 8081_37_exp

6 6

3 3 0 0 3 459 0 3 6 450 0 6

9 9 447 456 12 12

15 444 15 453

18 18 441 450 21 21 Y1_37_stat 8081_37_stat 447 438 24 24 444 435 812.1 681.6 27 27

44 30 432 30 406.1 1 340.9 438 33 429 33

36 36 435 426 0.1 0.1

39 39 432 423

42 42 429 420 45 45 426 417 48 48 423 414 51 51 420 411 54 54 417 408 57 57 414

405 60 60 411

402 63 63 408 399 66

66 405 396 69

69 402

393 72 72 399 390 75 75 396 387 78 78 393 81 384 81 390 84 381 84 387 87 378 87 384 90 375 90 381 93 372 93 378 96 369 96 375 99 366 99 372 102 363 102 369 105 360 105 366 108 357 108 363 111

354 111 360 114

351 114 357 117

348 117 354 120

345 120 351 123

342 123 348 126

339 126 345 129

336 129 342 132

333 132 339 135

330 135 336 138

138 327 333 141

324 141 330 144

321 144 327 147

318 147 324 150

315 150 321 153

312 153 318 156

156 159 309 315

159 162 306 312

162 165 303 309

165 168 300 306

168 171 297 303

174 171 300 294

174 177 291 297

180 177 288 294 183 180 291 285 186 183 288 282 189 186 285 279 192 189 282 276 195 192 279 273 198 195 276 270 201 198 273 267 204 201

270 207 264 204

267 210 261 207

264 213 210 258 216 213 261 255 219

216 258 252 222 219 255 225 249 222 228 225 252 231 228

2 234 246

249

3 243 4

1 246 240 3 243 237

240

2 237

C Chromosome pYV

●●● ● ● ● 0.2 0.3 ●●●● ● ●● ●● ●● ● ●● ● 0.2 ● ● ● ● 0.0 ● PC3(6.09%) ● ● PC3(7.88%) ● 0.1 ● ● ● ● ● ● ●● ● 0.0 −0.2 ● ● ● ● −0.1 ● ● ●● −0.4 ● ● ● 0.10 ● ● ● ●● −0.20 0.12 0.3 Groups: ● Groups: ● ● PC1(24.2%) 0.3 PC1(18.3%) ●● 0.2 8081_25_exp 8081_25_exp ●●● −0.15 0.14 ● ●● 8081_25_stat 0.2 8081_25_stat ● 0.1 8081_37_exp 8081_37_exp 0.16 0.0 8081_37_stat 0.1 8081_37_stat Y1_25_exp Y1_25_exp −0.1 Y1_25_stat −0.10 Y1_25_stat PC2(6.35%) 0.18 Y1_37_exp 0.0 PC2(13.3%) Y1_37_exp −0.2 Y1_37_stat Y1_37_stat YPIII_25_stat YPIII_25_stat −0.1 YPIII_37_stat YPIII_37_stat YPIII_25_exp YPIII_25_exp YPIII_37_exp YPIII_37_exp

Figure 4.1: Comparative RNA-seq workflow and global reports. (A) For comparative in vitro RNA- seq Y. enterocolitica serotype O:3 strain Y1 and serotype O:8 strain 8081v were grown in LB to exponential or stationary phase at 25°C or 37°C. Total RNA was isolated from bacterial cultures, processed for preparation of strand-specific barcoded cDNA libraries and sequenced. cDNA reads were separated in silico by mapping to 8081v and Y1 genomes. (B) Circos plot visualizing replicate mean averaged RPKM (Reads Per Kilobase transcript length per Million mapped reads) normalized expression values of in vitro RNA-seq data for the Y. enterocolitica 8081v chromosome (NC_008791.1) and its virulence plasmid (pYVO:8; NC_008800.1) and the Y. enterocolitica Y1 chromosome (CP030980) and its virulence plasmid (pYVO:3; CP030981). 25_exp: exponential phase 25°C; 37_exp: exponential phase 37°C; 25_stat: stationary phase 25°C; 37_stat: stationary phase 37°C. (C) Principal component analysis of mean centered and scaled rlog transformed read count values of the RNA-seq data for the Y. pseudotuberculosis YPIII (Nuss et al., 2015), Y. enterocolitica 8081v and the Y. enterocolitica Y1 genome sequences and their virulence plasmids. The analysis is based on the top 2802 core genes and 51 plasmid core genes based on row variance.

4 Publication 2

4.4.2 Genome-wide analysis of transcriptional start sites

To generate a comprehensive map of transcriptional start sites (TSSs) for Y. enterocolitica, we adopted a method used for Y. pseudotuberculosis (Nuss et al., 2015, 2017), in which four virulence-relevant growth conditions were used to monitor transcription activation. We categorized the identified transcription start as mTSS for mRNAs, lmTSS for leader less transcripts, sTSS and asTSS for the start site of small trans-acting regulatory RNAs and antisense RNAs (Fig. S4.2A). We identified 1299 mTSSs located upstream of the coding sequence for YeO:8 strain 8081v and 1076 mTSS for YeO:3 Y1 (Dataset S4.2). This revealed the global set of active gene promoters across the chromosome and the virulence plasmid of the species Y. enterocolitica for the first time. To validate the mTSSs, mTSSs of published genes/operons of Y. enterocolitica were compared with predicted mTSSs of the RNA-seq analysis. The vast majority of the small set of previously identified mTSSs are identical. As example, all mTSSs, which were previously identified within the ompF, ompX and ybtA upstream region in Y. enterocolitica (Anisimov et al., 2005; Gao et al., 2011) were also detected in our RNA-seq approach. We further determined the level of conservation of transcriptional organization between YeO:8 and YeO:3 by identifying common, but also strain-specific TSSs. The location of 882 mTSS identified for strain 8081v was conserved in Y1 (Dataset S4.2), and only 417 and 194 mTSSs were identified in YeO:8 8081v and in YeO:3 Y1, respectively. 23 mTSSs were identified on the virulence plasmid of Y1 (pYVO:3) and 23 mTSSs on the equivalent plasmid of 8081v (pYVO:8) of which 13 were identical in both plasmids (Dataset S4.2). Many of the strain-specific transcription sites are from hypothetical proteins, phages and mobile elements. Another reason for differences in the TSS is that several genes are not or only very little expressed in one strain under all tested growth conditions. As shown in Fig. 4.2A, the gene for the attachment and invasion locus ailA is significantly expressed in YeO:8 8081v with a transcript starting 59 nucleotides upstream of its translational start site, but not or only very little in YeO:3 Y1. Vice versa, the fliE gene is only expressed in Y1 (Fig. 4.2B). Moreover, for the expression of some genes different promoters are used. For instance, two start sites at position 80 and 35 upstream of the translational start site are observed for the sanA gene in YeO:3 Y1, but only the proximal TSS is used in YeO:8 8081v (Fig. 4.2C). Pairwise comparison of the promoter region revealed four nucleotide exchanges and one nucleotide insertion upstream of the more distal TSS in the YeO:8 8081v increasing the space between the putative -35 and -10 promoter region. It is very likely that this results in a

4 Publication 2

drastic loss of RNA polymerase binding and promoter activity. For some genes, such as rosB, different promoters are used under all tested growth conditions leading to 5'-UTRs with a significantly different length (247 nt in Y1 and 46 nt 8081v; Fig.4.2D). This indicated that there are many differences in the general promoter pattern, which could be either based on certain variations on the DNA sequence level of the regulatory region or on the distinct expression/function of regulatory proteins.

Figure 4.2: Comparative analysis of mRNA transcriptional start sites (TSSs) of Y1 and 8081v. (A) Visualization of RNA-seq based cDNA sequencing reads mapped to the 8081v and Y1 ail (A), fliE (B), sanA (C), and rosB gene locus (D) using the IGV genome browser. The transcriptional start sites are indicated by broken arrows. The promoter region of the sanA gene is given in panel (C). The -10 and -35 region of the upstream promoter of the sanA gene are underlined. The TSS identified in YeO:3 Y1 strain is given in red.

4.4.3 Global analysis of the promoter regions and architecture

To detect conserved promoter sequence motifs for canonical RNA polymerase (RpoD) binding sites, we used MEME (Bailey and Elkan, 1994) within the -10 region (position -15 to - 3) and the -35 region (position -45 to -25) assuming that RpoD is responsible for the majority

4 Publication 2

of transcription initiation as in Y. pseudotuberculosis and other related Enterobacteriaceae (Kim et al., 2012; Kröger et al., 2012; Nuss et al., 2015). Alignment of all identified TSSs and promoter sequences revealed that adenine is the most common initiating nucleotide (A>40%) and TAtaaT is the detected -10 Pribnow box region (Fig. S4.2B,C), which are very similar between YeO:3 and YeO:8, and highly homologous to Y. pseudotuberculosis (Nuss et al., 2015, 2017). In contrast to Y. pseudotuberculosis (with a -35 region of TTGC/A), but similar to some other pathogens (D’Arrigo et al., 2016; Dugar et al., 2013; Petersen et al., 2003; Sharma et al., 2010), no strong canonical -35 region could be identified, even when only 115 promoters with a high expression rate (> 100 reads) were included into the analysis. The identification of the TSSs further enabled us to define and analyze the untranslated regions (5’-UTRs). The majority of transcripts of YeO:3 Y1 and YeO:8 8081v possessed an 5'-UTR of 20-60 nt in length (Fig. S4.2D, Dataset S4.2) very similar to Y. pseudotuberculosis and other bacteria (Kröger et al., 2012, 2018; Sharma et al., 2010; Wurtzel et al., 2012). However, 141 and 182 5'-UTR's were longer than 150 nt, and a subset of 13 and 16 mRNAs had 5’-UTRs longer than 300 nt in YeO:3 Y1 and YeO:8 8081v (Datasets S4.2, S4.3). These long 5'-UTRs could include putative cis-regulatory RNA elements, such as RNA riboswitches and RNA thermometers, known to control transcription, translation initiation, and stability of mRNAs (Breaker, 2011; Henkin, 2008; Kortmann and Narberhaus, 2012; Serganov and Nudler, 2013). A more detailed inspection using RibEx riboswitch explorer and the Rfam database predicted 44 (YeO:8 8081v) and 32 (YeO:3 Y1) riboswitch-like elements (RLEs) or RNA thermometer among the long 5’-UTRs of which 25 were conserved in both strains (Datasets S4.2, S4.3). Several well-known RNA thermometer, e.g. the RNA thermometer of the pYV-encoded regulator lcrF and the FMN riboswitch of Yersiniae were identified (Böhme et al., 2012; Nuss et al., 2015; Righetti et al., 2016) , but also new interesting candidates for RNA thermometers (fimbrial mRNA fimA-6, and the T3SS component mRNAs yscH and yscD), and riboswitches (the crp mRNA for the cAMP regulatory protein) could be discovered. Moreover, 19 and 21 leaderless mRNAs with a 5’-UTR <10 nt (in Dataset S4.2) were identified in YeO:3 Y1 and YeO:8 8081v, which lack a classical ribosome-binding site and use an AUG as translation start codon. The latter was shown to be essential for stable ribosome-binding to leaderless transcripts (Brock et al., 2008). In total 24 and 17 TSSs were found to start downstream of the predicted start codon in strain Y1 and 8081v, likely due to an incorrect annotation of the translation initiation site.

4 Publication 2

Figure 4.3: Identification of ncRNAs of YeO:3 Y1 and YeO:8 8081v. Visualization of RNA-seq based cDNA sequencing reads of the Y. enterocolitica-specific sRNA Ysr212 (A, left panel) and the 8081v strain-specific sRNA Ysr109 (B, left panel) mapped to the 8081v and Y1 genome using the IGV genome browser. Differential expression of the trans-encoded sRNAs Ysr212 of strains Y1 and 8081v (A, right panel) and Ysr109 of strain 8081v (B, right panel) determined by qRT-PCR are shown.

4.4.4 The repertoire of Y. enterocolitica non-coding RNAs

As non-coding RNAs (ncRNAs) represent an important class of post-transcriptional regulators that modulate many cellular processes, including virulence, we used the Y. enterocolitica Y1 and 8081v transcriptomes to identify non-coding RNAs (ncRNAs). Using a conservative strategy applied to Y. pseudotuberculosis (Nuss et al., 2015), we could identify 262 (20%) and 486 (26.5%) ncRNAs in YeO:3 Y1 and YeO:8 8081v of which 119 and 204 were expressed from intergenic regions, so-called trans-encoded small RNAs (sRNAs), and 143 and 264 from the antisense strand of mRNAs (asRNAs) (Dataset S4.4). We listed them according to their location in relation to overlapping or nearby coding genes and named them Ysr(e) to distinguish them from the ncRNAs of other human pathogenic Yersiniae.

4 Publication 2

Figure 4.4: Comparison of the growth- and temperature-dependent regulons of YeO:3 Y1 and YeO:8 8081v. (A-C) Venn diagrams illustrating growth phase regulated (A) and temperature-regulated (B) protein-encoded genes of Y1 and 8081v. 4-way Venn diagrams illustrating core genes that are significantly upregulated (log2fc ≥ |2|; adjusted p-value ≤0.05) in 8081v (C) or upregulated in Y1 (D) under the indicated growth conditions. (E) Distribution of genes categorized based on their function that are upregulated in strain 8081v and Y1 at 25°C (left panel) or 37°C (right panel) during stationary phase growth.

4 Publication 2

A gene conservation analysis was performed by comparing the identified sRNAs in strain Y1 and 8081v with sRNAs identified in the human pathogenic yersiniae, Y. pestis and Y. pseudotuberculosis (Beauregard et al., 2013; Koo et al., 2011; Nuss et al., 2015, 2017; Qu et al., 2012; Schiano et al., 2014; Yan et al., 2013). We found that only 32 sRNAs are conserved among γ-proteobacteria and an additional 21 sRNAs have orthologs in the other two human pathogenic Yersiniae, Y. pestis and Y. pseudotuberculosis. 74 of the identified Y. enterocolitica-specific sRNAs are conserved between both strains and other members of the species (Dataset S4.4). This diversity of ncRNAs was also observed among Salmonellae, Campylobacter and Acinetobacter species (Dugar et al., 2013; Kröger et al., 2012, 2018) and may provide the bacteria with additional species-specific or even strain-specific regulatory functions important for bacterial fitness and virulence. We validated our ncRNA identification by qRT-PCR with sequence-specific probes designed to hybridize to species-conserved (e.g. Ysr021, Ysr060, Ysr138, Ysr143, Ysr212), and species-specific sRNAs (e.g. Ysr109) and confirmed condition-dependent expression of the identified ncRNA transcripts (Fig. 4.3, Fig. S4.3). Comparable with the RNA-seq data, the Ysr212 sRNA and the 8081v-specific Ysr109 sRNA were strongly growth-phase and temperature-regulated and were maximally expressed at 25°C under stationary phase in 8081v, whereas Ysr212 showed a more even expression under the tested environmental conditions in Y1. We further present a comprehensive expression landscape of all identified asRNAs in Dataset S4.4. It is anticipated that the asRNAs target the complementary mRNA, whereas the biological function and the interaction partners of the trans-encoded ncRNAs are not easy to predict and remain to be identified in future studies.

4.4.5 Monitoring of infection-relevant changes in YeO:3 and YeO:8 gene expression

To gain a better understanding of the genetic and molecular basis of the different host range and pathogenicity of YeO:3 and YeO:8 strains, we compared the expression of infection- linked genes between the YeO:3 and YeO:8 strain. To do so, we first defined the core genome of both strains (3347 genes) and profiled the entire transcriptional landscape of YeO:3 Y1 and YeO:8 8081v grown at exponential and stationary phase at 25°C and 37°C. Comparative RNA-seq analysis was performed using DESeq2 from triplicate experiments to identify genes that are differentially regulated by at least 4-fold (P-value ≤ 0.05) in response to growth phase or temperature (Datasets S4.5, S4.6). The RNA-Seq data visualized as

4 Publication 2

circos plots show consistent expression global profiles for all conditions, with uniformly high and low transcript abundance (Fig. 4.1B). Despite the overall high average nucleotide identity of 97% between YeO:3 Y1 and YeO:8 8081v (Fig. S4.1), the bacterial expression profiles of the chromosomal genes of both strains are distinct (Fig. 4.1B). Some genomic regions have a similar high (sydD-lpxDAB region) and low (thiM-mdaB region) expression pattern, whereas others exhibit a more distinct transcription profile (mtlADR-pckA region). Moreover, transcript abundance varies considerably during stationary phase, whereas the expression profiles of the exponentially grown cultures are much more alike between both strains (Fig. 4.1C). As expected, the expression pattern of both strains is significantly different from Y. pseudotuberculosis YPIII (86% sequence identity, Fig. S4.1) under all tested conditions (Fig. 4.1C) (Nuss et al., 2015).

Global differences in the expression profile of YeO:3 and YeO:8 in response to temperature The overall change of the expression profiles in response to growth is comparable, between the YeO:3 Y1 (30%) and the YeO:8 8081v strain (43%) (Fig. 4.4A). As expected many genes implicated in protein translation (e.g. ribosome and tRNA synthesis), cell division (murD, murC, mraY, ddl, mrdB/rodA, bolA) and starvation control (rssB, fadB-2, fadI, fadH, psiE) are growth-phase-dependent in both strains at both temperatures. Of the known virulence- relevant genes, several are only under growth phase control at 25°C (e.g. the catalase gene katA, the urease cluster ureABCDE) or at 37°C (e.g. fimbrial gene fimD-1) in both strains (Datasets S4.5, S4.6). Compared to growth phase control, the temperature-responsive regulons of YeO:3 Y1 and YeO:8 8081v are much more different. Observed expression changes in response to temperature are more prominent in YeO:8 8081v than in YeO:3 Y1 (Fig. 4.1C, Fig. 4.4A,B). Venn diagrams illustrate that considerably more genes are temperature-regulated (≥4-fold; p- value ≤0.05) in YeO:8 strain 8081v (1153 genes, 26% of the genome) compared to YeO:3 strain Y1 (320 genes, 7.4% of the genome), (Fig. 4.4A-D, Datasets S4.5, S4.6). This divergence is considered to be indicative for a change in the lifestyle or niche of both serotypes, in which differences in the expression pattern is a consequence of a transition from environmental ubiquity, including rodents and insect vectors, to specialization in enteric infection of animals such as pigs with an average body temperature of 40°C (Bottone, 1997; Fredriksson-Ahomaa et al., 2006a; Gürtler et al., 2005; Jones et al., 2003; Reuter et al., 2014). In fact, 581 mRNAs are more abundant in 8081v under at least one of the tested conditions, of which 292 (about 50%) are more expressed at 37°C. This is clearly different in

4 Publication 2

the serotype O:3 strain Y1. 650 mRNA transcripts are ≥ 4-fold higher expressed in Y1, of which 75% are more abundant at 37°C in particular during stationary phase (Fig. 4.4C-E). This set of genes includes many metabolic and cell physiology genes, indicating that the overall expression profile of the serotype O:3 strain has shifted towards higher temperatures (Fig. 4.4D,E, Datasets S4.5, S4.6). The molecular mechanisms responsible for the differences in the temperature and growth phase regulons are still unclear. However, several global regulators (e.g. the stationary phase sigma factor RpoS, the carbon storage regulator CsrA) as well as more specific transcription factors, that are differentially expressed under certain growth conditions in Y1 and 8081v, could contribute to this process (Datasets S4.7, S4.8). In contrast to the chromosome, the expression pattern of the virulence plasmid, encoding the Ysc type III secretion system (T3SS) and the antiphagocytic effectors called Yops, is comparable in both strains. The majority of T3SS/yop genes are thermo-induced in both strains to a similar level, independent of the growth phase (Fig. 4.1C, Datasets S4.6, S4.7), suggesting that the most prominent signals triggering ysc/yop virulence gene expression are very similar in both serotypes. To explicitly unravel differences in the regulatory networks controlling expression of virulence-relevant genes in YeO:3 and YeO:8 strains, we used our RNA-seq approach to screen for transcripts that differed significantly in their abundance between Y1 and 8081v under the different growth conditions or are coordinately regulated genes in response to temperature and nutrient limitation. We visualized the expression values of selected fitness- and virulence-linked genes between both strains in a heat map (Fig. 4.5A). To validate our analysis, DESeq2-estimated fold change responses for selected bacterial transcripts were confirmed by quantitative RT-PCR (Fig. S4.4).

Differential expression of metabolic functions indicates a distinct availability of essential nutrients In total, only 24/64 mRNAs were consistently enriched (4-fold/2-fold), whereas 29/86 mRNAs were consistently depleted in YeO:3 strain Y1 compared to YeO:8 8081v under all tested conditions (Datasets S4.7, S4.8). This set of transcripts includes the manZY mRNA encoding a PTS system, responsible for the transport of mannose, fructose and sorbose family sugars, which is more than 15-fold upregulated in Y1 (Fig. 4.5). Moreover, components of ion and nutrient transporters (yphF, ybbA-1, glpF-2, nupC-1, putP, npr) are more expressed in Y1, whereas others, a component of an iron uptake system (fhuD-2), and the

4 Publication 2

ascorbate/cellobiose PTS system genes (ulaABC) are more expressed in 8081v under all tested conditions. These metabolic functions may represent important fitness-relevant traits allowing a better colonization of the bacteria in their preferred hosts. A major difference was also observed for multiple metabolic genes, e.g. several enzyme genes encoding enzymes involved in amino acid (gadA, adiA, ilvAHM, cysM, metB, selA) and carbohydrate metabolism (ackA, pflA, pflB, fumC, mak_1, pgl_2, budA, yiaC, yjgM) are more expressed in Y1, whereas others, i.e. genes of the glycolysis-pyruvate-TCA cycle, are more expressed in 8081v (gdhA, NA_557/serralysin, ansAB, poxB, ppsA, actP, acs, sucA, fbp, pckA) (Fig. 4.5). Overall upregulation of pyruvate-acetyl-CoA TCA node enzymes in 8081v was particularly evident at 25°C under stationary phase, indicating that the primary carbon metabolism has adjusted to a more frequent environmental life-style, in contrast to Y1 which is a predominant mammalian-associated pathogen. Since the number of genes that are differently expressed under all tested conditions is relatively low, we hypothesized that the phenotypic differences that distinguish the serotypes are also caused by variations that modulate transcription of selected virulence genes and/or virulence-relevant fitness genes under certain growth/infection conditions. In fact, many transport and metabolic functions are only differentially expressed under one of the tested conditions. Examples are the cys transport and metabolic genes (cysAWTP, metB, cysIJCNDG2M2T) and the hemin-uptake hmuVUTSBP operon which are much more induced in YeO:3 Y1 at 25°C during exponential phase growth, and the ast operon (astEBDA) for amino acid metabolism that is mainly induced in YeO:8 at 25°C during stationary phase. In contrast, the maltose uptake system (lamB-1, malS, malZ-2, malMKEFG) is significantly more induced in YeO:8 at 25°C during exponential growth (Fig. 4.5A, Datasets S4.7, S4.8). Relative expression of some metabolic functions even changes, when the bacteria switch from the exponential to stationary phase. For instance, the fructose- specific PTS system (fruABK) is significantly more upregulated in YeO:3 Y1 at 25°C during exponential phase, whereas during stationary phase, it is much higher expressed in YeO:8 strain 8081, whereas the magnesium transporter mgtBC genes are more induced during stationary phase and less during exponential phase in Y1 (Fig. 4.5A, Datasets S4.7, S4.8). The different expression pattern of metabolic functions is likely to reflect the availability of nutrients in the intestine of the prefered hosts depending on the competing microbiota and the host diet.

4 Publication 2

Figure 4.5: Bacterial global gene expression analysis of YeO:3 Y1 and YeO:8 8081v uncovers strain-specific metabolic and stress adaptations. (A) Heat map of selected bacterial transcripts related to metabolic and stress adaptation functions, which are enriched (red) or depleted (blue) in strain YeO:3 Y1 compared to YeO:8 8081v. Values represent the log2 fold change of indicated conditions (adjusted p-value ≤0.05). (B) Central carbon metabolism of Y. enterocolitica. Significant changes of the transcriptomic pattern between Y1 and 8081v grown at 25°C during stationary phase are indicated. Enriched transcripts in Y1 are given in red, and enriched transcripts in 8081v are indicated in blue.

4 Publication 2

Comparative analysis reveals distinct mechanisms that may allow better survival in a given host Another major factor that determines the success of an infection in different hosts includes differences in the bacteria themselves, e.g. their ability to survive a given host environment. A striking difference between the analyzed strains is also that some genes important for the pH resistance of bacteria are only induced or more expressed in YeO:3 strain Y1. This set includes genes of the amino acid decarboxylase/antiporter systems, depending on arginine (AdiA/AdiC) or glutamate GadC/GadA) system, and the urease production and urea transporter genes ureABCEFGD and yut (Fig. 4.5A, Datasets S4.7, S4.8). These acidic + + resistance systems bind H ions to form CO2 or nitrate (NH4 ) (Lund et al., 2014). Moreover, a strongly pH-dependent sodium/proton antiporter gene nhaA and the nhaR activator gene are more than 100-fold induced during stationary phase in Y1 (Fig. 4.5A, S4.5). An increased pH resistance could be advantageous for the infection of humans and larger animals (pigs/boars) in which the mean residence time in their more voluminous stomach is likely to be prolonged compared to rodents and hares. Besides these alterations, other fitness-relevant gene expression differences were revealed between the different serotypes. The cold shock protein genes (cspB, cspC) are generally more induced in YeO:8 8081v, whereas the transcripts of the universal stress genes uspA and uspB, the carbon starvation gene cstA2 and the phoH gene for a putative phosphate starvation protein are more expressed in YeO:3 Y1 (Fig. S4.5A, Datasets S4.7, S4.8). Moreover, the flagella and chemotaxis genes (flhDC, motA, cheB-trs/cheD, fliZABCDFHIKLOR, flgKGCMN, flhB) are only expressed in YeO:8 8081v at 25°C during stationary phase, but not in Y1, consistent with a previous observation of our group (Uliczka et al., 2011a). This phenotype could be linked to the expression of the autoinducer AI-2 transport and degradation system (lsrACDBFG), which was shown to modulate biofilm formation, chemotaxis, motility and attachment to host cells (Pereira et al., 2013). The lsrACDBFG operon is mainly expressed at 25°C during stationary phase, condition under which the lsrRK mRNA, encoding the equivalent repressor LsrR and the regulator LsrK, is more than 4-fold less expressed (Fig. S4.5).

Differential expression of colonization, dissemination and toxin genes Several genes have been identified that are important for Yersinia virulence, host colonization and antimicrobial resistance. Our data set offers a detailed and comparative molecular analysis of the transcript abundance of well and less characterized

4 Publication 2

Y. enterocolitica virulence factors. Fimbrial and non-fimbrial adhesins are critical components in Y. enterocolitica virulence as they are required for efficient host cell binding to colonize the intestinal tract and subepithelial tissue. Several adhesin genes are differentially regulated between YeO:3 Y1 and YeO:8 8081v. Prominent examples are the fimbrial clusters mrfJHGF-papCD-mrfB-smfA2, fimA5-NA_360-fimD3-NA_361-NA_362, fimA6-fimC4-fimD-4- NA_538/6-NA_539/5, which are more expressed in YeO:8 8081v, whereas the PsaA antigen/fimbriae and the fimD2-fimC2 genes are higher induced in YeO:3 Y1 (Fig. 4.6, Datasets S4.7, S4.8). In addition, transcripts for the afimbrial adhesins, such as the outer membrane adhesin YapH and the primary cell adhesion and invasion factor InvA, are significantly more abundant in Y1 (Fig. 4.6, Datasets S4.7, S4.8). The latter observation is consistent with our previous study demonstrating an increased synthesis of InvA in porcine and human isolates of YeO:3 strains due to a constitutive promoter of an IS1667 element integrated into the invA promoter region and overall higher expression levels of the transcriptional activator RovA of invA (Uliczka et al., 2011a). In contrast, higher amounts of the mRNAs of the attachment and invasion locus AilA, the homologous protein AilD/OmpX, the InvA-type adhesin InvB/Ifp and the virulence plasmid-encoded adhesin YadA were detected in 8081v. This likely mirrors the selectivity/preference of the serotype O:3 and O:8 strains for certain host cells expressing the adhesin-specific cellular receptor. Alteration of the pathogen-host cell interactions will modulate the colonization and/or dissemination behaviour of the bacteria and may promote a serotype-specific preference for different hosts, e.g. humans, pigs/boars, hares or rodents. Many of these colonization factors are strongly controlled in response to temperature and growth phase in YeO:8 strain 8081v, but are more equally expressed in YeO:3 strain Y1. The most prominent difference was observed at 25°C during stationary phase in which the fimbrial cluster mRNA (mrfJHGF-papCD-mrfB-smfA2, fimA5-NA_360-fimD3-NA_361, fimA6-fimC4-fimD-4-NA_538/6-NA_539/5) is significantly more abundant in YeO:8 strain 8081v (Fig. 4.6, Datasets S4.7, S4.8). Expression of these adhesion structures at moderate temperatures may be important to colonizes certain environmental reservoirs and/or prime the bacteria to allow immediate and efficient colonization of the intestinal epithelium upon host entry. Another major factor that affects host specificity includes differences in the bacteria's ability to evade a given host immune response. Notably, two effector proteins, YopT and YopJ/YopP, which are injected into neutrophils and macrophages to perturb host innate immune responses, are more expressed in YeO:8 strain 881v (Fig. 4.6, Datasets S4.7, S4.8). YopT, a cysteine protease targeting the small GTPases Rac1, RhoA, Cdc42, and

4 Publication 2

RhoG, is implicated in the disruption of the actin cytoskeleton to contribute to the inhibition of phagocytosis. YopP/YopJ interferes with multiple IKKb-/NFkB- and MAPK-signalling components to inhibit pro-inflammatory cytokine and chemokine production, and activates caspase-1 and the maturation of IL-18 and IL1b to induce immune cell death (Pan et al., 2014). This suggests that a higher activity inhibiting and eliminating host phagocytes is required or advantageous for 8081v to survive in its preferred hosts. Conversely, the outer membrane protease Pla-2 and one of the most important virulence factors of Y. enterocolitica, the heat-stable enterotoxin A (YstA), is significantly more expressed in YeO:3 strain Y1 (Fig. 4.6, Datasets S4.7, S4.8), suggesting a significantly higher toxicity of this strain.

Fold changes [log2] of selected virulence genes

ystA invA ureA ureB 5 ureC ureE ureG ureD yut sodB 0 pla2 yapH fimD−2 fimC−2 rovA slyA psaA −5 psaC_1 psaE shlB hlyB rovM lrhA pecT yapE −10 fimA−2 yopK/yopQ mrfF mrfJ mrfG mrfB ppdA papD papC YEY1_02566/YE8081_02721 fimD−4 fimD−3 fimC−4 ompX/ailD ailA sodC rtxB rtxD_1 rtxE_2 yopP yopJ fimA−6 smfA−2 yadA yopT YEY1_02938|YE8081_03040 ifp invB fimA−5 25°C, exp 25°C, stat 37°C, exp 37°C, stat

YeO:8 8081v YeO:3 Y1

Figure 4.6: DifferentiallyFigure 6 Schmühl regulated et al. 2018 virulence functions between strains YeO:3 Y1 and YeO:8 8081v. Heatmaps of transcripts encoding virulence-related genes which are enriched (red) and depleted (blue) in strain YeO:3 Y1 compared to YeO:8 8081v. Values represent the log2 fold change of indicated conditions (adjusted p-value ≤0.05).

4 Publication 2

Figure 4.7: Analysis of ystA expression in Y. enterocolitica. (A) RNA was isolated from three independent cultures of different Y. enterocolitica serotype O:3, O:8 and O:9 isolates grown to stationary growth phase at 25°C. Expression of ystA relative to Y. enterocolitica Y1 was determined using qRT-PCR. qRT-PCR was performed in technical duplicates with DNA-free total RNA (primers are listed in Table S4.1). The 5S rRNA gene was used for normalization and relative gene expression changes were calculated according to Pfaffl 2001 (Pfaffl, 2001). The data represent the mean ± SEM of the relative expression from three independent biological replicates performed in triplicates. (B) The scheme illustrates the constructed plasmid-encoded ystA-lacZ translational fusion harboring different portion of the ystA promoter region. The numbers indicate the positions relative to the TSS (i) used for the cloning of the fusions, or (ii) to indicate the identified promoter. (C) The ystA-lacZ translational fusions harboring the 5'UTR of the ystA gene of Y1 or 8081v were transformed either in Y1 or into 8081v. β-galactosidase activity was determined. The data represent the mean ± SEM of the fold change (end/start) from three independent biological replicates performed in triplicates and were analyzed with Student’s t-test. The stars indicate the results that differed significantly from those of the wildtype harboring the identical reporter plasmid; **: P<0.01, ***: P<0.001.

4 Publication 2

4.4.6 Differential expression of the ystA toxin gene

The enterotoxin YstA is one of the most important and reliable virulence markers of Y. enterocolitica. It strongly influences Yersinia virulence and is a major causative agent of secretory diarrhea. In one study all 89 of 89 pathogenic and none of 51 non-pathogenic Y. enterocolitica isolates contained ystA-homologous genes (Delor et al., 1990). Moreover, rabbits infected with an ystA+ strain suffered from diarrhea, rapidly lost weight and most died, whereas rabbits infected with the ystA mutant showed no disease symptoms, and rapidly disappeared from the feces (Delor and Cornelis, 1992). The mechanism of YstA action is based on guanylate cyclase activation, which results in increased cGMP levels in enterocytes and extracellular liquids in the intestines (Inoue et al., 1983; Revell and Miller, 2000). It further stimulates the intracellular inositol triphosphate (IP3) levels that interacts with the IP3 receptor and mobilizes intracellular calcium in intestinal epithelial cells (Saha et al., 2009). Our comparative RNA-seq analysis revealed that the gene of the YstA toxin (ystA) is significantly higher expressed in the serotype O:3 strain Y1 compared to serotype O:8 strain 8081v (Fig. 4.6, Datasets S4.7, S4.8). Next, we compared the abundance of the ystA transcript at 37°C during stationary phase between Y1 and a group of ystA-positive clinical isolates of Y. enterocolitica representing different biotypes isolated from distinct geographical regions of the world at different time points. All isolates of the serotype O:8 and O:9, as well as 'older' isolates of YeO:3 (collected before 2007) exhibited varying, but in general very similar low expression levels of the toxin (Fig. 4.7A). In contrast, most of the isolates obtained over the past 10 years produced higher levels of the ystA transcript (Fig. 4.7A). It is possible, that 'older' isolates switched ystA expression to a silent state as described for some isolates (Mikulskis et al., 1994). Alternatively, the more recent strains might have acquired an additional mutation leading to an increase of ystA gene transcription or ystA mRNA stability. As the ystA promoter region of the YeO:3 strains Y11 and 1203 with low ystA transcript levels (Fig. S4.6) is 100% identical to that of YeO:3 Y1 with high ystA mRNA amounts, we assumed that distinct expression levels are the result of differences in a trans-encoded factor. In fact, transcriptional ystA-lacZ fusions harboring the entire ystA promoter region of Y1 and 8081v (position -582 to + 11 with respect to the translational start site) are both highly expressed in YeO:3 strain Y1, but were fully repressed in YeO:8 strain 8081v (Fig. 4.7B-C). Different deletions of the ystA promoter region resulted in a progressive increase in ystA transcription in YeO:8 8081v (Fig. 4.7B-C), suggesting that an additional negative regulatory

4 Publication 2

protein represses ystA expression, but this silencing is relieved in YeO:3 Y1. The close inspection of the ystA upstream region revealed high AT abundance and the occurrence of long polyAT-rich stretches upstream of the transcriptional start site overlapping the identified promoter in this and a previous study (Fig. S4.6, Dataset S4.2) (Mikulskis et al., 1994). This indicates a high DNA flexibility and characterizes the predominant binding and nucleation sites of the global nucleoid-associated regulator H-NS (Atlung and Ingmer, 1997; Schröder and Wagner, 2002). Interaction of H-NS with these sites leads to polymerization and the formation of higher order nucleoprotein complexes resulting in the repression of the target promoter downstream (Dame et al., 2005). To investigate a potential role of H-NS in silencing of ystA in 8081v, which seems eliminated in Y1, we measured expression of the ystA-lacZ in Y1 in the presence of a hns+ plasmid and found that epitopic expression of the hns gene leads to a strong repression of ystA, very similar to what is seen in 8081v (Fig. 4.8A). A similar influence was observed for YmoA (Fig. 4.8B), an H-NS homologue, that interacts directly with H-NS and forms a repression complex silencing a subset of H-NS controlled virulence genes (Cathelyn et al., 2006). This strongly indicated that H-NS/YmoA- mediated repression of ystA in 8081v is relieved in Y1, potentially by an activator protein that counteracts H-NS function. One obvious candidate is RovA. RovA was shown to counteract H-NS and YmoA-regulated genes in Yersinia including invA and psaA (Cathelyn et al., 2006; Ellison and Miller, 2006; Heroven et al., 2004), which are both upregulated in Y1 compared to 8081v (Fig. 4.6, Datasets S4.6, S4.7). In fact, rovA expression was found to be much more induced in YeO:3 Y1 compared to YeO:8 at all tested conditions, but in particular during stationary phase, in which the ystA transcript is mostly increased (Fig. 4.6, Datasets S4.6, S4.7). This is in full agreement with a previous study of our group showing that the amount of RovA in YeO:8 8081v (and the YeO:3 strain Y11 with low ystA transcript levels), is lower compared to YeO:3 (Uliczka et al., 2011a) This is caused by a P98S substitution in RovA. This amino acid exchange renders the regulator less susceptible to proteolysis and results in a more efficient autoactivation of its transcription (Uliczka et al., 2011a). We tested whether a mutation in rovA and overexpression of rovA influences expression of the ystAO:3-lacZ and ystAO:8-lacZ fusions and found that RovA is able to enhance ystA expression in Y1 (Fig. 4.8C). However, although this effect is significant, the overall influence is rather low, suggesting that an additional regulatory factor might contribute to ystA upregulation in Y1.

4 Publication 2

A Influence of H-NS on ystA-lacZ B Influence of YmoA on ystA-lacZ

**** **** *** *** ** ***

80 80 y y t i t i v i v i t ) t ) ) ) n ac i n 60

ac 60 i

.m .m g ase g ase m id m id ( s ( 40 l/

s 40 l/ o o o t o t m m ac μ l ac l μ ( a ( a 20

G 20 G - -

0 0 + pV phns pV phns+ pV phns+ pV phns+ pV pymoA+ pV pymoA+ pV pymoA+ pV pymoA+

P (Y1) P (8081v) P (Y1) ystA ystA ystA PystA (8081v) PystA (Y1) PystA (8081v) PystA (Y1) PystA (8081v)

Y1 8081v Y1 8081v C Influence of RovA on ystA-lacZ

*** 80 ***

y t i

v *** i ) t ) 60 *** n i ac

*m g ase m id ( s

l/ 40 o o t m ac μ l ( a G

- 20

+ +

A A lacZ v v - -lacZ o o ) ) ) ) r r A-lacZA-lacZ p p (O:3 (O:3 (O:8 Z Z A A (O:8 yst yst lac lac yst yst - ) - ) rovA rovA A A Y1 Y1 1 1 (O:3 (O:8 Y Y yst yst Y1 Y1

Figure 4.8: Influence of H-NS, YmoA and RovA on ystA expression of Y. enterocolitica strain Y1. Plasmids encoding the hns gene (A), the ymoA gene (B) or the rovA gene (C) were transformed into YeO:3 strain Y1 or the isogenic rovA mutants carrying a ystA-lacZ fusion construct with the entire ystA promoter region of the ystA gene of Y1 or 8081v. The strains were grown to stationary phase at 25°C, and β-galactosidase activity was determined. The data represent the mean ± SEM of the fold changeFigure (end/start) 8 Schmühl from et threeal. 2018 independent biological replicates performed in triplicates and were analyzed with Student’s t-test. The stars indicate the results that differed significantly from those of the wildtype harboring the identical reporter plasmid; **: P<0.01, ***: P<0.001.

4 Publication 2

4.5 Conclusions

The ability of pathogenic bacteria to reprogram their fitness- and virulence-related traits can adapt them to other environmental reservoirs and host. This can lead to unexpected outbreaks and epidemics in distinct host species populations, and is thus a global public and veterinary health concern. To obtain information about the molecular basis of host tropism mainly population genomic studies have primarily been applied. They have provided the core genome of the genus and led to the identification of specific point mutations (SNPs), gene gain, gene loss and genome rearrangement events that influence host adaptation pathways and specificity in Yersinia and other bacterial pathogens (Batzilla et al., 2011; Hammarlöf et al., 2018; McNally et al., 2016; Reuter et al., 2014; Sheppard et al., 2013; Viana et al., 2015). Of the functions that were altered as different Y. enterocolitica lineages evolved and adapted to new host niches is the cell adhesion and invasion factor InvA. In the highly mouse-virulent phylogroup 2/serotype O:8 strains, invA is strongly temperature-regulated and predominantly transcribed at 25°C during stationary phase. However, in phylogroups 3/serotype O:3 strains, which show limited pathogenesis in mice, but have become the dominant isolate found in pig reservoirs and cases of human disease, an IS1667 element integrated into the invA promoter. This created a new promoter and an additional binding site for the RovA activator that ensures constitutive expression of the invasin gene (Uliczka and Dersch, 2012; Uliczka et al., 2011a). The upregulation of invA enabled a more efficient colonization of porcine tissue compared to other phylogroups (Schaake et al., 2014), suggesting that this is the primary event that led tot he enhanced virulence observed in recent isolates from the phylogroup 3/O:3 strains. In this study, we followed a different approach and compared the first primary transcriptomes of Y. enterocolitica using strains 8081v and Y1, representing the phylogroups 2 and 3, to determine the transcriptional variability in the response to infection-relevant conditions. This revealed strain-specific promoter usage, sRNA repertoires and uncovered different transcriptional outputs that are also likely to facilitate adaptation to different host niches and impact pathogenesis. Integrating of the comparative dRNA-seq data from both strains under four different growth conditions improved the annotation accuracy and allowed us to determine 1299 and 1076 TSSs of mRNAs in 8081v and Y1 of which the majority (1213 and 1043) belong to the core genome and are conserved among both strains. However, also many examples of strain-specific promoter usage were identified, and although some promoters are highly conserved the respective genes are not necessarily expressed at the same level by Y1 and 8081v. One prominent example is the ystA gene which is strongly

4 Publication 2

induced in Y1, in particular during stationary phase, but not in 8081 and other 'older' serotype O:3 isolates with an identical promoter region. This illustrates that comparative transcriptomics is an excellent approach to discover differences in the functional output from genomes, which cannot be directly inferred from closely related DNA sequences. Overall, our high-resolution transcriptome map discovered major differences in the transcription pattern between the phylogroup 2 and 3 strain, in particular of the temperature- responsive regulon. Multiple fitness- and virulence-relevant genes are controlled in response to temperature and often expressed at a higher level at 25°C in the serotype O:8 strain 8081v, whereas no significant or a much less extensive thermal response is observed in the homologous genes in the serotype O:3 isolate Y1. This most likely reflects differences in the life-style of the bacteria and points to a recent study proposing an ecological separation with certain niche-adapted pathogenic lineages of Y. enterocolitica (Reuter et al., 2015). Although all phylogroups of Y. enterocolitica can be isolated from the intestinal tract of cattle sheep and pigs, serotype O:8/phylogroup 2 strains are rarely isolated from humans and livestock, and have a higher level of virulence in mouse infection models. Moreover, analysis of the core and accessory genes and the gene flow across the phylogroups suggest that different phylogroups are ecologically separated and do not seem to share common niches (Reuter et al., 2012, 2015). Observed genetic and transcriptional differences can be adaptive and lead to niche expansion/separation. A variety of pathoadaptive alterations were identified which can affect (i) host cell binding, colonization dissemination and host tissue tropism, (ii) their ability to evade or overcome immune mechanisms, (iii) availability to survive stresses, (iv) uptake and utilization of essential nutrients for growth and (v) virulence regulation. All these features are important for virulence and determine host specificity/tropism (Pan et al., 2014). The most striking difference has been determined for the acid resistance genes, the adhesins and the enterotoxin YstA. The ystA mRNA is much more abundant in Y1 compared to 8081v. This indicates a much higher toxicity of Y1. However, exotoxin function is linked with the ability of the pathogen to adhere to the intestinal epithelial layer, i.e. the bacteria require a colonization factor that promotes tight interaction with intestinal epithelial cells for the onset of diarrhea. Some Y. enterocolitica fimbriae and the afimbrial adhesin invasin (InvA) which are more strongly expressed in YeO:3 Y1 at body temperature are likely candidates, as they guarantee that the serotype O:3 strains are much better colonizer of the pig intestine, than the serotype O:8 (Schaake et al., 2013, 2014). Enhanced expression of the YstA toxin and improved adherence of this strain, which might facilitate absorption of the enterotoxin, may explain the

4 Publication 2

strong diarrhea of the patients of these isolates. How this combination of adhesion factors and the toxin impacts pathogenesis needs to be characterized in future studies. However, it is notable, that a similar cocktail of virulence factors leading to more efficient aggregative adherence by the newly emerged Escherichia coli serotype O104:H4 was shown to account for the increased uptake of Shigatoxin toxin into the systemic circulation, resulting in high rates of the hemolytic-uremic syndrome (Navarro-Garcia, 2014).

4.6 Data availability

The complete Y1 genome sequence was deposited in NCBI under accession numbers CP030980 (chromosome YEY1_1) and CP030981 (plasmid YEY1_2). All high-throughput short read data, gene expression quantification information and DESeq2 result list for all comparisons are deposited at the Gene Expression Omnibus (GEO) database with the accession no. GSE119404. A complete list of the TSSs, antisense and trans-encoded sRNAs is provided in datasets S4.2 and S4.3. The comparative transcriptome analyses are given in datasets S4.4–S4.8. The annotation of global TSSs, reannotated ORFs, assigned locus tags, trans- and cis-encoded sRNAs are provided as supplementary files (Files S4.1– S4.15).

4.7 Supplementary data

4.7.1 Supplementary tables

Table S4.1: Bacterial strains, plasmids and oligonucleotides Bacterial Strains Bacterial Strains Description Source & Reference E. coli DH10β F - endA1 recA1 galE15 galK16 nupG rpsL galU Durfee et al., 2008 deoR ΔlacX74 Φ80dlacZΔM15 araD139Δ(ara, leu)7697 mcrA Δ(mrr-hsdRMS-mcrBC)λ- CC118λpir F Δ(ara, leu) 7697 Δ(lacZ)74 Δ(phoA)20 araD139 Manoil and Beckwith, galE galK thi rpsE rpoB arfE 1986 am recA1, λpir Y. enterocolitica Y1 O:3, Biotype 4, BfR.Nr. Y1/07, human isolate from Uliczka et al., 2011a feces of a diarrhea patient, 2007 8081v O:8, Biotype 1B, USA Portnoy et al., 1981

4 Publication 2

Y34 O:3, Biotype 4, BfR.Nr. Y34/07, human isolate from Uliczka et al., 2011a feces of a diarrhea patient, 2007 08-01985 O:3, Biotype 4, RKI-Nr. 08-01985, human isolate RKI, F. Uliczka from feces of a diarrhea patient, 2008 FBI_033 O:3, Biotype 4, FBI_033 TiHo, F. Uliczka

Y59 O:3, Biotype 4, BfR.Nr. Y54/07 (Y59) Uliczka et al., 2011a

Y71 O:3, Biotype 4, BfR.Nr. Y8/08 (Y71) Uliczka et al., 2011a

Y11 O:3, Biotype 4, Y11, sequenced by DSMZ:13030 Uliczka et al., 2011a

6471/76 O:3, 6471/76, wildtype patient isolate M. Skurnik

FBI_03778 O:3 Biotype 4, FBI_03778 TiHo, P. Valentin- Weigand C32M1 O:3, isolate from Europe, A. Mellado M. Skurnik

JH 5700/84 O:3, isolate from Hamburg, 1984, J. Heesemann M. Skurnik

JH 1131/84 O:3, isolate from Hamburg, 1984, J. Heesemann M. Skurnik

JD E675 O:3, human isolate from North America, before 1980, M. Skurnik J. Devenish

S.T.8204 O:3, isolate from Canada, before 1985, S. Toma M. Skurnik

80016 O:3, Biotype 4, human stool isolate, 2006, MaKeRa M. Skurnik project, Helsinki, Finland 84053 O:3, Biotype 4, human stool isolate, 2006, MaKeRa M. Skurnik project, Helsinki, Finland

1870/73 O:3, isolate from Turku, Finland, 1973 M. Skurnik

1150/73 O:3, isolate from Turku, Finland, 1973 M. Skurnik

37 O:3, API 0114520, isolate from Oulu, Finland, 1985 M. Skurnik

GKp774 O:3, isolate from Norway 1975-1985, M. Skurnik G. Kapperud

GK21603 O:3, isolate from Norway 1975-1985, M. Skurnik G. Kapperud

E675 O:3, isolate from Europe, A. Mellado M. Skurnik

RBC36M3 O:3, isolate, 1983, B. Perry M. Skurnik

1203 O:3, Biotype 4 A. McNally

Y18 O:8, Biotype 1B, BfR.Nr.79/90 (Y18) Uliczka et al., 2011a

Y19 O:8, Biotype 1B, BfR.Nr. Ye12 (Y19) Uliczka et al., 2011a

Y49 O:8, Biotype 1B, BfR.Nr. 39/91 (Y49) Uliczka et al., 2011a

4 Publication 2

07-00628 O:9, Biotype 3, RKI-Nr. 07-00628 RKI, F. Uliczka

Y24 O:9, Biotype 3, BfR.Nr. 2012/79 (Y24) Uliczka et al., 2011a

Y31 O:9, Biotype 3, BfR.Nr. 495/88 (Y31) Uliczka et al., 2011a

Y48 O:9, Biotype 3, BfR.Nr. 7/91 (Y48) Uliczka et al., 2011a

21202 O:9, Biotype 2 A. McNally

5603 O:9, Biotype 3 A. McNally

YE12 Y1 ΔrovA Uliczka et al., 2011a

Plasmids Plasmid Description Source & Reference pAKH71 pACYC184, ymoA+, CmR Böhme et al., 2012 pAKH74 pACYC184, hns+, CmR Heroven et al., 2004 pHT109 pZA31, rovA+, CmR H. Tran-Winkler pFU55 Cloning vector for lacZ (translational fusions), Uliczka et al., 2011b pSC101, *KanR pCS68 pFU55, ystA(O:8)-lacZ, , fusion of PystA (position -- This study 582 to +11) to lacZ, pSC101, *KanR pCS63 pFU55, ystA(O:3)-lacZ, fusion of PystA (position -203 This study to +11) to lacZ, pSC101, *KanR pCS68 pFU55, ystA(O:8)-lacZ, fusion of PystA (position -389 This study to +11) to lacZ, pSC101, *KanR pCS70 pFU55, ystA(O:8)-lacZ, fusion of PystA (position -203 This study to +11) to lacZ, pSC101, *KanR pCS71 pFU55, ystA(O:3)-lacZ, fusion of PystA (position - 582 This study to +11) to lacZ, pSC101, *KanR pCS72 pFU55, ystA(O:3)-lacZ, fusion of PystA (position - 389 This study to +11) to lacZ, pSC101, *KanR

Oligonucleotides Name Sequence (5’- 3’ orientation) Description/Target Sequence Oligonucleotides used for cloning VIII9 GCGGCGCTCGAGTCTGGTAACGAAAAGAG forward primer for ystA upstream region (-582) of Y. enterocolitica O:3 and O:8 harboring a XhoI site VIII009 GCGGCGCCTCGAGTCTGGTAACGAAGAG cloning of ystA 5’UTR (-582 bp) of Y. enterocolitica O:3 and O:8 harboring a XhoI site VIII010 GCGGCGCCTCGAGAGTTATTATTCACAACAAAGG forward primer for ystA (O:3) upstream region (-389) harboring a XhoI site VIII011 GCGGCGCCTCGAGAGATATAAACTATGATTAAATTA forward primer for ystA (O:3) GC upstream region (-203) harboring XhoI site VIII014 GCGGCGCCTCGAGAGTTATTATTCACAATAAAG forward primer for ystA (O:8) upstream region (-389) harboring a XhoI site VIII015 GCGGCGCCTCGAGATATAAACTATGATTAAATTAGT forward primer for ystA (O:8)

4 Publication 2

G upstream region (-203) harboring XhoI site VIII016 GCGGCGCGCTAGCCTATCTTTTTCATAGATCCTCC reverse primer ystA (O:3) coding region (+11) harboring NheI site VIII019 GCGGCGCGCTAGCTACCTTTTTCATAGAACCTCC reverse primer ystA (O:8) coding region +11) harboring NheI site Oligonucleotides used for qRT-PCR VIII22 GATAGTTTTTGTTCTTGTG forward primer qRT for ystA, Y. enterocolitica O:3 VIII23 CTAGCAGCCAGCACAC forward primer qRT for ystA, Y. enterocolitica O:8 VIII40 CTAGCAACCCGCACAG reverse primer qRT for ystA, Y. enterocolitica O:3 and O:8 III95 ACCCCGCTGAACATAATGAG forward primer qRT for invA of Y. enterocolitica O:3 and O:8 III96 TGCCGCGTCATTTACCATTG reverse primer qRT for invA of Y. enterocolitica O:3 and O:8 III182 CCGGTGGTTTGCACGGCGT forward primer qRT for gyrB of Y. enterocolitica O:3 and O:8 III183 CACCACTTTCAATGGTGCC reverse primer qRT for gyrB of Y. enterocolitica O:3 and O:8 VIII41 AACGCAAAGCGCGTGGC forward primer qRT for ureA of Y. enterocolitica O:3 and O:8 VIII42 GGAATCAGATCAGCCACC reverse primer qRT for ureA of Y. enterocolitica O:3 and O:8 VIII57 TGAAACACTTACGCACCC forward primer qRT for metR of Y. enterocolitica O:3 and O:8 VIII58 GTTGTGAAACGTAACGGC reverse primer qRT for metR of Y. enterocolitica O:3 and O:8 VIII83 CAATTCAATATAAATGAATTTGG forward primer qRT for smfA of Y. enterocolitica O:3 and O:8 VIII95 TGCTGGTCTGGTATTAGG forward primer qRT for invA of Y. enterocolitica O:3 and O:8 VIII96 CGCCATTTTGCAGTGCC reverse primer qRT for smfA of Y. enterocolitica O:3 and O:8 VIII101 GTAGCAATGGCAGCAAGC forward primer qRT for fimA-6 of Y. enterocolitica O:3 and O:8 VIII102 AACCTGATACACCCAGAG reverse primer qRT for fimA-6 of Y. enterocolitica O:3 and O:8 VIII105 AAGAACTGATTGTTGCCAC forward primer qRT for glnH of Y. enterocolitica O:3 and O:8 VIII106 CCAGTGCCAGATCCACG reverse primer qRT for glnH of Y. enterocolitica O:3 and O:8 VIII121 GTGGTACGAGGCGAAGG forward primer qRT for astC of Y. enterocolitica O:3 and O:8 VIII122 GCCAGTCGCAACACCGG reverse primer qRT for astC of Y. enterocolitica O:3 and O:8 VIII123 AACGATGTTCATTTACGC forward primer qRT for leuO of Y. enterocolitica O:3 and O:8 VIII124 ACCTCGACCATAACGCAC reverse primer qRT for leuO of Y. enterocolitica O:3 and O:8 VIII441 TATCTATCTATCTATCTATCTATC reverse primer qRT-PCR Ysr(e)021 in YeO:3 VIII442 ATTATGGATTATTGTTCTC forward primer qRT-PCR Ysr(e)021 in YeO:8

4 Publication 2

VIII454 CCAAGAGTTTTCTGGCAC forward primer qRT-PCR Ysr(e)060 VIII455 GAACTTAATTTCTATTGGCG reverse primer qRT-PCR Ysr(e)060 VIII469 TGTTATGCAATAGTCATGC forward primer qRT-PCR Ysr(e)109 VIII470 AAACCCCGGCGAAAACC reverse primer qRT-PCR Ysr(e)109 VIII480 TCGACACGGCGCTGTG forward primer qRT-PCR Ysr(e)143 in YeO:8 VIII481 CGCGGGAGGCAGATAATACG reverse primer qRT-PCR Ysr(e)143 VIII493 ATGGCGGCTGAGTGCTG forward primer qRT-PCR Ysr(e)212 VIII494 GTTCGTCATGCGCCACC reverse primer qRT-PCR Ysr(e)212

BfR: Federal Agency for Risk Assessment Germany RKI: Robert Koch Institute, Germany TiHo: University of Veterinary Medicine Hanover, Germany

4 Publication 2

4.7.2. Supplementary Figures

Figure S4.1: Average nucleotide identity of the YeO:3 strain Y1 with other Y. enterocolitica and Y. pseudotuberculosis strains. Nucleotide sequence of the YeO:3 strain Y1 was compared with the Y. enterocolitica serotype O:3 strains Y11 and 1203, and Y. pseudotuberculosis strains YPIII and IP32953. The average nucleotide identity with other strains is indicated in percentage using the pyANI (https://github.com/widdowquinn/pyani) and MUMmer (Kutz et al., 2004) software.

4 Publication 2

Figure S4.2: Global identification of mRNA transcriptional start sites (TSSs). (A) Schematic overview of the identified TSS: mTSS for mRNAs, lmTSS for leader less transcripts, sTSS and asTSS for the start site of small trans-acting regulatory RNAs and antisense RNAs. (B) Sequence conservation at the TSSs. Sequence logo computed from 1299 unaligned TSS of YeO:8 strain 8081v and 1076 unaligned TSS regions (TSS is located at position +1) showing nucleotide conservation around the TSSs. The initial nucleotides of transcripts (position +1 to +3) are dominated by purines. (C) Detected conserved sequence motifs in the -10 promoter region (Pribnow Box). (D) The distribution and frequency of the length of 5’-UTRs is given for all mRNAs of Y1 and 8081v, which start upstream of the annotated TSS. More than 40% of all 5’-UTRs are 20-60 nt in length.

4 Publication 2

Figure S4.3: Identification of ncRNAs of YeO:3 Y1 and YeO:8 8081v. Visualization of RNA-seq based cDNA sequencing reads of the Y. enterocolitica sRNA Ysr021 (A, left panel), Ysr060 (B, left panel), and Ysr143 (C, left panel) mapped to the 8081v and Y1 using the IGV genome browser. Differential expression of the trans-encoded sRNAs Ysr012 (A, right panel), Ysr060 (B, right panel), and Ysr143 (C, right panel) determined by qRT-PCR are shown. For qRT-PCR three independent cultures of YeO:3 Y1 and YeO:8 8081v were grown in LB medium to exponential or stationary growth phase at 25°C or 37°C. qRT-PCR was performed in technical duplicates with DNA-free total RNA (primers are listed in Table S1). The gyrB gene was used for normalization and relative gene expression changes were calculated according to Pfaffl 2001. Black bars: real-time qRT-PCR for Y1; grew bars: real-time qRT-PCR for 8081v.

4 Publication 2

Figure S4.4: Comparison of gene expression changes obtained by RNA-seq and real-time qRT- PCR. Relative gene expression changes were examined for selected genes in response to temperature or growth phase. Three independent cultures of YeO:3 Y1 and YeO:8 8081v were grown in LB medium to exponential or stationary growth phase at 25°C or 37°C. qRT-PCR was performed in technical duplicates with DNA-free total RNA (primers are listed in Table S1). The gyrB gene was used for normalization and relative gene expression changes were calculated according to Pfaffl 2001. Black bars: real-time qRT-PCR; grew bars: RNA-seq.

4 Publication 2

Figure S4.5: Gene expression analysis of stress adaption genes and regulators of YeO:3 and YeO:8. Heatmaps of transcripts encoding stress adaptation genes (A) or regulators (B) which are enriched (red) and depleted (blue) in strain YeO:3 Y1 compared to YeO:8 8081v. Values represent the log2 fold change of indicated conditions (adjusted p-value ≤0.05).

4 Publication 2

Figure S4.6: Promoter region of the ystA gene in different Y. enterocolitica strains. A sequence comparison is shown of the ystA promoter region of the YeO:3 strains Y1, Y11, 1203 and the YeO:8 strain 8081v. The coding region is marked with a blue box, and the AT-rich regions with a red box. The identified transcriptional start site is indicated by a broken arrow, and the putative -10 and -35 region are marked with a bar.

4 Publication 2

4.8 Acknowledgement

We thank Dr. M. Fenner for helpful discussions. We also thank Bettina Elxnat, Nicole

Heyer, Tanja Krause and Simone Severitt for excellent technical assistance.

4.9 Funding

This work was supported by the Helmholtz Association and the Leibniz Association, C. Schmuehl was supported within the Ph.D. program ’Animal and Zoonotic Infections’ of the University of Veterinary Medicine Hannover by a Lichtenberg fellowship 'Niedersächsische Ministerium für Wissenschaft und Kultur (MWK)'. P. Dersch is supported by the German Center of Infection Research (DZIF).

4.10 References

Abreu-Goodger, C., and Merino, E. (2005). RibEx: a web server for locating riboswitches and other conserved bacterial regulatory elements. Nucleic Acids Res. 33, W690-692.

Altschul, S. F., Madden, T. L., Schäffer, A.A., Zhang, J., Zhang, Z., Miller, W., and Lipman, D.J. (1997). Gapped BLAST and PSI-BLAST: a new generation of protein database search programs. Nucleic Acids Res. 25, 3389–3402.

Amman, F., Wolfinger, M. T., Lorenz, R., Hofacker, I.L., Stadler, P.F., and Findeiß, S. (2014). TSSAR: TSS annotation regime for dRNA-seq data. BMC Bioinformatics 15, 89.

Anders, S., Pyl, P.T., and Huber, W. (2015). HTSeq--a Python framework to work with high- throughput sequencing data. Bioinformatics 31, 166–169.

Anisimov, R., Brem, D., Heesemann, J., and Rakin, A. (2005). Molecular mechanism of YbtA- mediated transcriptional regulation of divergent overlapping promoters ybtA and irp6 of Yersinia enterocolitica. FEMS Microbiol. Lett. 250, 27–32.

Aronesty, E. (2011). http://code.google.com/p/ea-utils

Atlung, T., and Ingmer, H. (1997). H-NS: a modulator of environmentally regulated gene expression. Mol. Microbiol. 24, 7–17.

Bailey, T.L., and Elkan, C. (1994). Fitting a mixture model by expectation maximization to discover motifs in biopolymers. Proc Int Conf Intell Syst Mol Biol 2, 28–36.

4 Publication 2

Beauregard, A., Smith, E.A., Petrone, B.L., Singh, N., Karch, C., McDonough, K.A., and Wade, J.T. (2013). Identification and characterization of small RNAs in Yersinia pestis. RNA Biol 10, 397– 405.

Böhme, K., Steinmann, R., Kortmann, J., Seekircher, S., Heroven, A.K., Berger, E., Pisano, F., Thiermann, T., Wolf-Watz, H., Narberhaus, F., et al. (2012). Concerted actions of a thermo-labile regulator and a unique intergenic RNA thermosensor control Yersinia virulence. PLoS Pathog. 8, e1002518.

Bottone, E.J. (1997). Yersinia enterocolitica: the charisma continues. Clin. Microbiol. Rev. 10, 257– 276.

Breaker, R.R. (2011). Prospects for riboswitch discovery and analysis. Mol. Cell 43, 867–879.

Brock, J. E., Pourshahian, S., Giliberti, J., Limbach, P.A., and Janssen, G.R. (2008). Ribosomes bind leaderless mRNA in Escherichia coli through recognition of their 5’-terminal AUG. RNA 14, 2159– 2169.

Cathelyn, J. S., Crosby, S. D., Lathem, W.W., Goldman, W.E., and Miller, V.L. (2006). RovA, a global regulator of Yersinia pestis, specifically required for bubonic plague. Proc. Natl. Acad. Sci. U.S.A. 103, 13514–13519.

Crooks, G. E., Hon, G., Chandonia, J.-M., and Brenner, S.E. (2004). WebLogo: a sequence logo generator. Genome Res. 14, 1188–1190.

Dame, R. T., Luijsterburg, M.S., Krin, E., Bertin, P.N., Wagner, R., and Wuite, G.J.L. (2005). DNA bridging: a property shared among H-NS-like proteins. J. Bacteriol. 187, 1845–1848.

D’Arrigo, I., Bojanovič, K., Yang, X., Holm Rau, M., and Long, K.S. (2016). Genome-wide mapping of transcription start sites yields novel insights into the primary transcriptome of Pseudomonas putida. Environ. Microbiol. 18, 3466–3481.

Delor, I., and Cornelis, G.R. (1992). Role of Yersinia enterocolitica Yst toxin in experimental infection of young rabbits. Infect. Immun. 60, 4269–4277.

Delor, I., Kaeckenbeeck, A., Wauters, G., and Cornelis, G.R. (1990). Nucleotide sequence of yst, the Yersinia enterocolitica gene encoding the heat-stable enterotoxin, and prevalence of the gene among pathogenic and nonpathogenic yersiniae. Infect. Immun. 58, 2983–2988.

Dugar, G., Herbig, A., Förstner, K. U., Heidrich, N., Reinhardt, R., Nieselt, K., and Sharma, C.M. (2013). High-resolution transcriptome maps reveal strain-specific regulatory features of multiple Campylobacter jejuni isolates. PLoS Genet. 9, e1003495.

Durfee, T., Nelson, R., Baldwin, S., Plunkett, G., 3rd, Burland, V., Mau, B., Petrosino, J.F., Qin, X., Muzny, D.M., Ayele, M. et al. (2008) The complete genome sequence of Escherichia coli DH10B: insights into the biology of a laboratory workhorse. J Bacteriol, 190, 2597-2606

4 Publication 2

ECDC. (2015). http://www.ecdc.europa.eu Vol. 2015.

Ellison, D.W., and Miller, V.L. (2006). H-NS represses inv transcription in Yersinia enterocolitica through competition with RovA and interaction with YmoA. J. Bacteriol. 188, 5101–5112.

Fredriksson-Ahomaa, M., Stolle, A., Siitonen, A., and Korkeala, H. (2006a). Sporadic human Yersinia enterocolitica infections caused by bioserotype 4/O : 3 originate mainly from pigs. J. Med. Microbiol. 55, 747–749.

Fredriksson-Ahomaa, M., Stolle, A., and Korkeala, H. (2006b). Molecular epidemiology of Yersinia enterocolitica infections. FEMS Immunol. Med. Microbiol. 47, 315–329.

Gao, H., Zhang, Y., Han, Y., Yang, L., Liu, X., Guo, Z., Tan, Y., Huang, X., Zhou, D., and Yang, R. (2011). Phenotypic and transcriptional analysis of the osmotic regulator OmpR in Yersinia pestis. BMC Microbiol. 11, 39.

Gürtler, M., Alter, T., Kasimir, S., Linnebur, M., and Fehlhaber, K. (2005). Prevalence of Yersinia enterocolitica in fattening pigs. J. Food Prot. 68, 850–854.

Hammarlöf, D.L., Kröger, C., Owen, S.V., Canals, R., Lacharme-Lora, L., Wenner, N., Schager, A.E., Wells, T.J., Henderson, I.R., Wigley, P., et al. (2018). Role of a single noncoding nucleotide in the evolution of an epidemic African clade of Salmonella. Proc. Natl. Acad. Sci. U.S.A. 115, E2614– E2623.

Henkin, T.M. (2008). Riboswitch RNAs: using RNA to sense cellular metabolism. Genes Dev. 22, 3383–3390.

Inoue, T., Okamoto, K., Moriyama, T., Takahashi, T., Shimizu, K., and Miyama, A. (1983). Effect of Yersinia enterocolitica ST on cyclic guanosine 3’,5’-monophosphate levels in mouse intestines and cultured cells. Microbiol. Immunol. 27, 159–166.

Jones, T.F., Buckingham, S.C., Bopp, C.A., Ribot, E., and Schaffner, W. (2003). From pig to pacifier: chitterling-associated yersiniosis outbreak among black infants. Emerging Infect. Dis. 9, 1007–1009.

Kalvari, I., Nawrocki, E.P., Argasinska, J., Quinones-Olvera, N., Finn, R.D., Bateman, A., and Petrov, A.I. (2018a). Non-Coding RNA Analysis Using the Rfam Database. Curr Protoc Bioinformatics 62, e51.

Kalvari, I., Argasinska, J., Quinones-Olvera, N., Nawrocki, E.P., Rivas, E., Eddy, S.R., Bateman, A., Finn, R.D., and Petrov, A.I. (2018b). Rfam 13.0: shifting to a genome-centric resource for non- coding RNA families. Nucleic Acids Res. 46, D335–D342.

Kim, D., Hong, J.S.-J., Qiu, Y., Nagarajan, H., Seo, J.-H., Cho, B.-K., Tsai, S.-F., and Palsson, B.Ø. (2012). Comparative analysis of regulatory elements between Escherichia coli and Klebsiella pneumoniae by genome-wide transcription start site profiling. PLoS Genet. 8, e1002867.

4 Publication 2

Koboldt, D.C., Zhang, Q., Larson, D.E., Shen, D., McLellan, M.D., Lin, L., Miller, C.A., Mardis, E.R., Ding, L., and Wilson, R.K. (2012). VarScan 2: somatic mutation and copy number alteration discovery in cancer by exome sequencing. Genome Res. 22, 568–576.

Kortmann, J., and Narberhaus, F. (2012). Bacterial RNA thermometers: molecular zippers and switches. Nat. Rev. Microbiol. 10, 255–265.

Kröger, C., Dillon, S.C., Cameron, A.D.S., Papenfort, K., Sivasankaran, S.K., Hokamp, K., Chao, Y., Sittka, A., Hébrard, M., Händler, K., et al. (2012). The transcriptional landscape and small RNAs of Salmonella enterica serovar Typhimurium. Proc. Natl. Acad. Sci. U.S.A. 109, E1277-1286.

Kröger, C., MacKenzie, K.D., Alshabib, E.Y., Kirzinger, M.W.B., Suchan, D.M., Chao, T.-C., Akulova, V., Miranda-CasoLuengo, A.A., Monzon, V.A., Conway, T., et al. (2018). The primary transcriptome, small RNAs and regulation of antimicrobial resistance in Acinetobacter baumannii ATCC 17978. Nucleic Acids Res.

Langmead, B., and Salzberg, S.L. (2012). Fast gapped-read alignment with Bowtie 2. Nat. Methods 9, 357–359.

Li, H., and Durbin, R. (2009). Fast and accurate short read alignment with Burrows-Wheeler transform. Bioinformatics 25, 1754–1760.

Li, H., Handsaker, B., Wysoker, A., Fennell, T., Ruan, J., Homer, N., Marth, G., Abecasis, G., Durbin, R., and 1000 Genome Project Data Processing Subgroup (2009). The Sequence Alignment/Map format and SAMtools. Bioinformatics 25, 2078–2079.

Love, M.I., Huber, W., and Anders, S. (2014). Moderated estimation of fold change and dispersion for RNA-seq data with DESeq2. Genome Biol. 15, 550.

Lund, P., Tramonti, A., and De Biase, D. (2014). Coping with low pH: molecular strategies in neutralophilic bacteria. FEMS Microbiol. Rev. 38, 1091–1125.

Manoil, C. and Beckwith, J. (1986) A genetic approach to analyzing membrane protein topology. Science, 233, 1403-1408.

McNally, A., Thomson, N.R., Reuter, S., and Wren, B.W. (2016). “Add, stir and reduce”: Yersinia spp. as model bacteria for pathogen evolution. Nat. Rev. Microbiol. 14, 177–190.

Mikulskis, A.V., Delor, I., Thi, V.H., and Cornelis, G.R. (1994). Regulation of the Yersinia enterocolitica enterotoxin Yst gene. Influence of growth phase, temperature, osmolarity, pH and bacterial host factors. Mol. Microbiol. 14, 905–915.

Miller, J.H. (1992) A short course in bacterial genetic: a laboratory manual and handbook for Escherichia coli and related bacteria. Cold Spring Habor, New York.

4 Publication 2

Navarro-Garcia, F. (2014). Escherichia coli O104:H4 Pathogenesis: an Enteroaggregative E. coli/Shiga Toxin-Producing E. coli Explosive Cocktail of High Virulence. Microbiol Spectr 2.

Nuss, A.M., Beckstette, M., Pimenova, M., Schmühl, C., Opitz, W., Pisano, F., Heroven, A.K., and Dersch, P. (2017). Tissue dual RNA-seq allows fast discovery of infection-specific functions and riboregulators shaping host-pathogen transcriptomes. Proc. Natl. Acad. Sci. U.S.A. 114, E791–E800.

Pan, X., Yang, Y., and Zhang, J.-R. (2014). Molecular basis of host specificity in human pathogenic bacteria. Emerg Microbes Infect 3, e23.

Pereira, C.S., Thompson, J.A., and Xavier, K.B. (2013). AI-2-mediated signalling in bacteria. FEMS Microbiol. Rev. 37, 156–181.

Petersen, L., Larsen, T.S., Ussery, D.W., On, S.L.W., and Krogh, A. (2003). RpoD promoters in Campylobacter jejuni exhibit a strong periodic signal instead of a -35 box. J. Mol. Biol. 326, 1361– 1372.

Pfaffl, M.W. (2001). A new mathematical model for relative quantification in real-time RT-PCR. Nucleic Acids Res. 29, e45.

Portnoy, D.A., Moseley, S.L. and Falkow, S. (1981) Characterization of plasmids and plasmid- associated determinants of Yersinia enterocolitica pathogenesis. Infect Immun, 31, 775-782.

Qu, Y., Bi, L., Ji, X., Deng, Z., Zhang, H., Yan, Y., Wang, M., Li, A., Huang, X., Yang, R., et al. (2012). Identification by cDNA cloning of abundant sRNAs in a human-avirulent Yersinia pestis strain grown under five different growth conditions. Future Microbiol 7, 535–547.

Rakin, A., Schneider, L., and Podladchikova, O. (2012). Hunger for iron: the alternative siderophore iron scavenging systems in highly virulent Yersinia. Front Cell Infect Microbiol 2, 151.

Reuter, S., Thomson, N.R., and McNally, A. (2012). Evolutionary dynamics of the Yersinia enterocolitica complex. Adv. Exp. Med. Biol. 954, 15–22.

Reuter, S., Corander, J., de Been, M., Harris, S., Cheng, L., Hall, M., Thomson, N.R., and McNally, A. (2015). Directional gene flow and ecological separation in Yersinia enterocolitica. Microb Genom 1, e000030.

4 Publication 2

Revell, P.A., and Miller, V.L. (2000). A chromosomally encoded regulator is required for expression of the Yersinia enterocolitica inv gene and for virulence. Mol. Microbiol. 35, 677–685.

Righetti, F., Nuss, A.M., Twittenhoff, C., Beele, S., Urban, K., Will, S., Bernhart, S.H., Stadler, P.F., Dersch, P., and Narberhaus, F. (2016). Temperature-responsive in vitro RNA structurome of Yersinia pseudotuberculosis. Proc. Natl. Acad. Sci. U.S.A. 113, 7237–7242.

Rosner, B.M., Stark, K., and Werber, D. (2010). Epidemiology of reported Yersinia enterocolitica infections in Germany, 2001-2008. BMC Public Health 10, 337.

Saha, S., Chowdhury, P., Mazumdar, A., Pal, A., Das, P., and Chakrabarti, M.K. (2009). Role of Yersinia enterocolitica heat-stable enterotoxin (Y-STa) on differential regulation of nuclear and cytosolic calcium signaling in rat intestinal epithelial cells. Cell Biol. Toxicol. 25, 297–308.

Sambrook, J. (2001) Molecular Cloning: A Laboratory Manual,. Cold Spring Harbor Laboratories, Cold Spring Harbor, NY.

Schiano, C.A., Koo, J.T., Schipma, M.J., Caulfield, A.J., Jafari, N., and Lathem, W.W. (2014). Genome-wide analysis of small RNAs expressed by Yersinia pestis identifies a regulator of the Yop- Ysc type III secretion system. J. Bacteriol. 196, 1659–1670.

Schlüter, J.-P., Reinkensmeier, J., Barnett, M.J., Lang, C., Krol, E., Giegerich, R., Long, S.R., and Becker, A. (2013). Global mapping of transcription start sites and promoter motifs in the symbiotic α-proteobacterium Sinorhizobium meliloti 1021. BMC Genomics 14, 156.

Schröder, O., and Wagner, R. (2002). The bacterial regulatory protein H-NS--a versatile modulator of nucleic acid structures. Biol. Chem. 383, 945–960.

Seemann, T. (2014). Prokka: rapid prokaryotic genome annotation. Bioinformatics 30, 2068–2069.

Serganov, A., and Nudler, E. (2013). A decade of riboswitches. Cell 152, 17–24.

Sheppard, S.K., Didelot, X., Meric, G., Torralbo, A., Jolley, K.A., Kelly, D.J., Bentley, S.D., Maiden, M.C.J., Parkhill, J., and Falush, D. (2013). Genome-wide association study identifies vitamin B5 biosynthesis as a host specificity factor in Campylobacter. Proc. Natl. Acad. Sci. U.S.A. 110, 11923–11927.

4 Publication 2

Tauxe, R.V. (2002). Emerging foodborne pathogens. Int. J. Food Microbiol. 78, 31–41.

Uliczka, F., and Dersch, P. (2012). Unique virulence properties of Yersinia enterocolitica O:3. Adv. Exp. Med. Biol. 954, 281–287.

Uliczka, F., Pisano, F., Schaake, J., Stolz, T., Rohde, M., Fruth, A., Strauch, E., Skurnik, M., Batzilla, J., Rakin, A., Heeseman, J. and Dersch, P. (2011a). Unique cell adhesion and invasion properties of Yersinia enterocolitica O:3, the most frequent cause of human Yersiniosis. PLoS Pathog. 7, e1002117.

Uliczka, F., Pisano, F., Kochut, A., Opitz, W., Herbst, K., Stolz, T., and Dersch, P. (2011b). Monitoring of gene expression in bacteria during infections using an adaptable set of bioluminescent, fluorescent and colorigenic fusion vectors. PLoS ONE 6, e20425.

Viana, D., Comos, M., McAdam, P.R., Ward, M.J., Selva, L., Guinane, C.M., González-Muñoz, B.M., Tristan, A., Foster, S.J., Fitzgerald, J.R and Penadés, J. R. (2015). A single natural nucleotide mutation alters bacterial pathogen host tropism. Nat. Genet. 47, 361–366.

Wurtzel, O., Yoder-Himes, D.R., Han, K., Dandekar, A.A., Edelheit, S., Greenberg, E.P., Sorek, R., and Lory, S. (2012). The single-nucleotide resolution transcriptome of Pseudomonas aeruginosa grown in body temperature. PLoS Pathog. 8, e1002945.

Yan, Y., Su, S., Meng, X., Ji, X., Qu, Y., Liu, Z., Wang, X., Cui, Y., Deng, Z., Zhou, D., Jiang, W., Yang, R. and Han, Y. (2013). Determination of sRNA expressions by RNA-seq in Yersinia pestis grown in vitro and during infection. PLoS ONE 8, e74495.

5 Discussion

RNA-sequencing has been well established and improved in recent years. This method provides an unbiased high-throughput sequencing approach that can capture global transcriptional responses in a strand-specific manner. It has been successfully applied to several bacterial pathogens, such as Pseudomonas aeruginosa, Campylobacter jejuni, Vibrio cholerae and Yersinia pseudotuberculosis. (Butcher and Stintzi, 2013; Dötsch et al., 2012; Mandlik et al., 2011; Nuss et al., 2015). Here, RNA-sequencing was used to compare the transcriptional profiles of Y. enterocolitica serotypes O:8 (YeO:8) and O:3 (YeO:3). YeO:8 strain 8081v is a well characterized representative of this serotype, which most knowledge about Y. enterocolitica is derived from. YeO:3 strain Y1 is a recent human stool isolate from an outbreak in Germany. The strains were grown under different in vitro conditions (25°C and 37°C, exponential and stationary growth phase) to obtain sequencing samples. These conditions are commonly used to mimic different temperatures and nutritional environments that enteropathogenic Yersinia encounter during their lifecycle. Additionally, the transcriptional profile of Y. pseudotuberucolis strain IP32953 grown at these conditions was compared to bacteria from in vivo samples 3 days post infection from murine Peyer’s Patches. As a result, this study provides novel in-depth information on the transcriptional landscape and the repertoire of novel regulatory and sensory RNAs of enteropathogenic Yersinia as well as global maps of TSS for both Yersinia species.

5.1 Global Mapping of transcriptional start sites in Yersinia

Mapping of TSS revealed new genetic information of Y. entercolitica and Y. pseudotuberculosis such as the global locations of TSS, active promoters at the tested conditions, potential riboswitches and novel non-coding RNAs. This newly defined transcriptional landscape provides a solid foundation for future studies on transcriptional regulation processes. Although algorithmic approaches for the determination of TSS exist, there are several false positive or false negative results for the automatic annotation of TSS. Therefore, TSS

5 Discussion

identification is still performed or curated manually in most cases as has been done in the present study (Stazic and Voß, 2016). This study provides the first comprehensive mapping of TSS in Y. enterocolitica. At the same time, it allows the comparison of the transcriptional organization of two serotypes. For annotated ORFs, 1299 TSS were identified in YeO:8 8081v and 1076 in YeO:3 Y1 (Dataset S4.2). The numbers are comparable to the 1312 TSS assigned to CDS in Y. pseudotuberculosis IP32953 (Dataset S3.2). Previously, 1151 TSS were identified for Y. pseudotuberculosis YPIII (Nuss et al., 2015). TSS maps have also been created for several other pathogens. 1907 TSS were identified in Helicobacter pylori and 1873 in Salmonella (Kröger et al., 2012; Sharma et al., 2010). For C. jejuni strains the number of TSS varied between 1905 and 2167 (Dugar et al., 2013). Generally, the numbers of identified TSS in several species varies greatly (reviewed in Stazic and Voß, 2016). During an infection, pathogens encounter several forms of stress: change in pH, osmotic stress or exposure to reactive oxygen species. These are only some of the conditions that might provide a specific signal for the activation of additional promoters and they could reveal additional TSS in Yersinia. It was shown for Salmonella that the number of identified TSS was increased from 1873 to 3383 when 21 conditions, including stress conditions like acidic pH, high osmolarity, low oxygen or exposure to bile salts were applied to the bacterial cultures (Kröger et al., 2012, 2013). In Pseudomonas putida, more than 600 additional TSS were identified when the bacteria were grown in glucose compared to other carbon sources (D’Arrigo et al., 2016). More TSS could be possibly added to the maps of Y. enterocolitica and Y. pseudotuberculosis if special growth conditions or environmental signals would be taken into account. However, the TSS identified in this study are in agreement with previous reported TSS, which shows that the data presented here provide a valuable basis for further analysis, such as the analysis of TSS conservation, distribution of initiating nucleotides, detection of promoter motifs, 5’-UTR length distribution or the detection of riboswitches. In Y. enterocolitica 882 TSS are conserved between the two strains assessed in this study (Dataset S4.2). This accounts for 68% of the annotated TSS of YeO:8 and 83% of YeO:3. The comparison between several Salmonella strains revealed that most TSS were conserved (Hammarlöf et al., 2018). In Listeria 88% of the TSS are conserved (Wurtzel et al., 2012), but only 48% of TSS in Campylobacter (Dugar et al., 2013). These studies show the high variation in the amount of TSS conservation between different species. The initiating nucleotide in both Y. enterocolitica strains is an adenine in more than 40% of the cases; to a lesser extend a threonine or guanine may be found (Fig 4.S2). This discovery

100

5 Discussion

is similar to Y. pseudotuberculosis IP32953 as well as Y. pseudotuberculosis YPIII (Fig. 3.S3; Nuss et al., 2015). These findings correspond to what has been observed for other species (Kröger et al., 2012; Mendoza-Vargas et al., 2009). Differences between the Yersinia species are detected with regard to the promoter motif. The -10 region is very similar in all enteropathogenic Yersinia strains (Fig 3.S3 and 4.S2), while no conserved -35 region was found in Y. enterocolitica, in contrast to Y. pseudotuberculosis, Salmonella and E. coli (Figure 3.S3; Kröger et al., 2012). However, the lack of a conserved overrepresented -35 promotor motif is not unusual as it was also described for P. putida, H. pylori and C. jejuni (D’Arrigo et al., 2016; Dugar et al., 2013; Petersen et al., 2003; Sharma et al., 2010). 267 genes (17%) and 252 genes (16,8%) in YeO:8 and YeO:3 have at least one alternative TSS, such as sanA of YeO:3 (Fig. 4.2C; Dataset S4.2). Alternative TSSs are very common in bacteria. They are found for 15% of the genes in H. pylori (Sharma et al., 2010), 33% in C. jejuni (Dugar et al., 2013), 22% in Geobacter sulfurreducens (Qiu et al., 2010) and up to 60% in E. coli (Thomason et al., 2015). The fact that, for example, an alternative TSS was detected for sanA in YeO:3, but not in YeO:8, while no TSS was detected for sanA in Y. pseudotuberculosis, shows clear strain-specific variations, suggesting distinct regulatory patterns. The global TSS annotation further allowed the analysis of the length distribution of 5’ UTRs and the presence of potential regulatory elements, which are known to regulate the translation of the corresponding mRNA (Henkin, 2008; Kortmann and Narberhaus, 2012). Most genes in YeO:8, YeO:3 and Y. pseudotuberculosis IP32953 have a 5’UTR with a length between 20 nt and 60 nt (Fig. 3.S3 and 4.S2). This is also true for Y. pseudotuberculosis YPIII and other organisms such as P. aeroginousa, H. pylori, E. coli and Klebsiella (Kim et al., 2012; Nuss et al., 2015; Sharma et al., 2010; Wurtzel et al., 2012). One exception is P. putida, which shows a higher variation in the length of its 5’ UTRs (D’Arrigo et al., 2016). Again, strain-specific differences can be observed. The 5’UTR of rosB is 45 nt long in Y. pseudotuberculosis and in YeO:8. However, in YeO:3 an 247 nt long UTR for rosB was detected (Dataset S3.2 and S4.2). 5’UTRs can harbor regulatory elements that regulate transcriptional or posttranscriptional processes (Winkler and Breaker, 2005). Several potential riboswitches were detected in the long 5’UTRs of Y. pseudotuberculosis and Y. enterocolitica. Some were present in all three analyzed strains, like RLE0214 in the 5’UTR of crp or RLE0149 in the 5’UTR of deoC. A riboswitch in the 5’UTR of both Y. enterocolitica strains, but not in Y. pseudotuberculosis was

101

5 Discussion

detected for example for glnA, pepA or gapA. Other riboswitches were only detected in one of the tested strains, for example RLE0024 for oppA of YeO:8 or RLE0240 and RLE0310 for corA of YeO:3. This indicates different, in some cases even strain-specific regulatory mechanisms.

5.2 The Yersinia repertoire of non-coding RNAs

New regulatory mechanisms were discovered based on the data from RNA-sequencing experiments, many of them involved the activity of regulatory RNAs. This study provides detailed data about the presence of so far undiscovered regulatory RNAs in Y. enterocolitica and Y. pseudotuberculsis IP32953. The numbers of newly identified sRNA candidates varied greatly between both Y. enterocolitica strains: 208 sRNAs were identified in YeO:8 and 119 in YeO:3 (Dataset S4.4). The latter number is comparable to 132 sRNAs identified in Y. pseudotuberculosis IP32953 (Dataset S3.2), 78 sRNAs in Y. pseudotuberculosis YPIII (Nuss et al., 2015), 94 potential sRNAs in Clostridium difficile (Soutourina et al., 2013) and 104 sRNA candidates that were found in Y. pestis (Yan et al., 2013). The number of sRNAs in YeO:8 is almost twice as high, being more similar to 280 sRNAs that have been discovered in Salmonella (Kröger et al., 2013). These differences in sRNA number of YeO:8 and YeO:3 underline variations in the sRNA repertoire not only between species, but even within the same species. Some Y. enterocolitica sRNAs were identified in both strains as well as in further Yersinia species. Some discovered sRNAs in YeO:3 and YeO.8 were not conserved in Y. pseudotuberculosis or Y. pestis, but rather in Y. kristensii, Y. frederikensii or Y. rohdei, probably because these species are most closely related to Y. enterocolitica and therefore share a higher sequence homology (Reuter et al., 2014). However, many sRNAs are specific to one serotype or even the strain used. This is especially striking for the sRNAs identified for YeO:8 8081v (Dataset S4.4). High species or strain specificity of sRNAs has also been shown for H. pylori, Pseudomonoas aeruginosa and C. jejuni (Dugar et al., 2013; Sharma et al., 2010; Wurtzel et al., 2012). Most of the sRNAs detected in Y. pestis and Y. pseudotuberculosis were shown to be Yersinia specific, without homologues for example in E. coli or Salmonella (Beauregard et al., 2013; Koo et al., 2011; Nuss et al., 2015), similar to what was observed in the present study. It has been shown before, that there are only a

102

5 Discussion

few sRNAs conserved, even when different Yersinia species are compared (Beauregard et al., 2013; Koo et al., 2011). This indicates that closely related species or even strains have evolved distinct RNA-based regulatory networks. A reason for this might be specific regulatory mechanisms, which contribute to diverse expression phenotypes of the closely related bacterial strains. Another explanation might be that although no sequence homology was found, some sRNAs might adopt similar secondary structures. It was shown that the sRNA binding protein ProQ recognizes its targets based on the secondary structure of it’s target RNA rather than on the RNA sequence (Holmqvist et al., 2018). Moreover, for trans- encoded RNAs, a weak homology to their target mRNAs is sufficient for interaction, since they usually control several targets (Storz et al., 2011; Waters and Storz, 2009). Therefore, it is possible that some sRNAs could not be identified as the same trans-encoded RNA due to missing sequence homology, but still have the same function. This is further supported in the present study by the results obtained from the in vivo analysis Y. pseudotuberculosis. Deletion of a single sRNA did not impair colonization, likely because the mutation was compensated for during infection (Fig. S3.8). This compensation could be achieved by a high redundancy in sRNA function. Despite the high number of identified transcripts in various organisms, only a few sRNAs have been functionally characterized. It was even assumed that they might only represent noise due to unspecific transcription in bacteria (Georg and Hess, 2011; Raghavan et al., 2012; Thomason et al., 2015). However, some functions have been uncovered and linked to important virulence traits in recent years. Some RNAs were shown to be involved in fitness- relevant mechanisms. For Y. pseudotuberculosis it was shown that several sRNAs are part of the cAMP-CRP regulon, which is strongly temperature-dependent, suggesting that these sRNAs are involved in the coupling of metabolism and virulence (Nuss et al., 2015). RsaE, an sRNA in Staphylococcus aureus is involved in the regulation of several metabolic pathways, by regulating the expression of genes that are involved in the amino acid transport, cofactor synthesis, TCA cycle as well as lipid, purine and carbohydrate metabolism (Bohn et al., 2010; Geissmann et al., 2009). Furthermore, it was reported that the YstA toxin of Y. enterocolitica requires the sRNA chaperone Hfq for maximum expression (Nakao et al., 1995). This finding indicates that the expression of virulence factors such as YstA might be regulated by sRNAs. The fact that Hfq is necessary for full virulence in all pathogenic Yersinia species shows the importance of sRNAs for Yersinia infection (Geng et al., 2009; Kakoschke et al., 2014; Schiano et al., 2010). Therefore, the identification of novel sRNA

103

5 Discussion

candidates in Y. enterocolitica provides a solid basis for further investigations of these sRNAs, their regulatory mechanism and their molecular function. The in vivo analysis of Y. pseudotuberculosis supports the significance of sRNAs during host infection. Several sRNAs were found to be upregulated in the Peyer’s Patches. Due to functional redundancy, no effect could be determined for deletion mutants of single sRNAs. The loss of one sRNA can seemingly be compensated for by the bacterium. However, when several sRNAs were deleted simultaneously, an impairment in colonization of the Peyer’s Patches and mesenteric lymph nodes was observed, highlighting the importance of sRNAs during the cause of an infection.

The largest differences between the Yersinia strains can be observed for the numbers of detected antisense RNAs. 143 asRNAs have been identified for YeO:3, 264 for YeO:8 and only 45 for Y. pseudotuberculosis IP32953 (Datasets S3.2 and S4.4). Previous studies discovered 80 putative asRNAs in Y. pseudotuberculosis YPIII (Nuss et al., 2015). In YeO:3, the asRNAs have been assigned to 140 genes, meaning that 3.2 % of all genes have been associated with an antisense transcript. For YeO:8, asRNAs were assigned to 255 genes (5.9% of all genes). These numbers are relatively low compared to other organisms. The percentages of genes that have been associated with at least one antisense transcript ranges from 13% in Bacillus subtilis over 35% in Sinorhizobium meliloti and 46% in H. pylori to 49% in S. aureus (Lasa et al., 2011; Nicolas et al., 2012; Schlüter et al., 2013; Sharma et al., 2010). A reason for the low numbers of detected asRNAs in Yersinia could be that a rather stringent approach for the detection of ncRNAs was used, where only very clear candidates were added to the list of asRNAs. Another explanation could be the sequencing depth. It is possible that some asRNAs are described only at a low level and deeper sequencing would reveal more asRNAs in Yersinia species. However, since the numbers of sRNAs varies greatly between organisms, Yersinia might express lower amounts of asRNAs and preferably uses other regulatory mechanisms instead. Low conservation of antisense transcripts, as shown here for Yersinia, has previously been described for other pathogens such as Listeria, E. coli, Salmonella, C. jejuni (Dugar et al., 2013; Raghavan et al., 2012; Wurtzel et al., 2012). It was hypothesized that asRNAs create a rapid way for evolutionary fine tuning (Stazic and Voß, 2016). New asRNAs evolve easily during mutations of the DNA (Saberi et al., 2016). In the cyanobacterium Synechocystis sp. PCC6803, a single point mutation is sufficient to create an asRNA that affects the function of photosystem II of the bacterium (Kopf and Hess, 2015; Sakurai et al., 2012). This might also

104

5 Discussion

be the case for trans-encoded sRNAs. Therefore, the low conservation level of regulatory RNAs in enteropathogenic Yersinia does not necessarily conclude that they have no function or are falsely annotated. In fact, the low conservation could be an evolutionary mechanism of regulation, allowing a fast and sensitive adaption to new environments or host niches.

5.3 Changes in the gene expression profile

Rapid adaptation to new situations based on environmental clues is a crucial property for bacterial survival. Adaptation to changing environments is accompanied by reprogramming their regulatory network in order to activate genes that are essential for survival in the given environment while repressing unnecessary or potentially harmful genes (Pal et al., 2005).

During their life cycle, enteropathogenic Yersinia encounter varying environments, in which they have to ensure sufficient nutrient access to persist, compete with other microorganisms and avoid the host immune system. Changes in temperature and nutrient availability are known to be an important signal for gene expression adaptation in bacteria, affecting a large regulon. For example, in P. aeroginosa 427 genes are temperature regulated (Wurtzel et al., 2012). Similar to Yersinia, P. aeroginosa is able to live in a variety of host organisms. This study revealed that a large set of genes (43% of the total protein coding genes) is regulated in a growth phase dependent manner in YeO:8. Less, but still plenty genes (30%) are growth phase regulated in YeO:3. Temperature regulated genes are accounting for 26% and 7.4 % of all genes (Datasets S4.5 and S4.6). The sizes of the growth phase and temperature regulon in YeO:3 are more similar to those observed for Y. pseudotuberculosis YPIII where 23% of all genes are expressed in response to growth phase and 7.6 % in response to temperature (Nuss et al., 2015). Both Yersinia species share the growth phase- dependent expression of genes involved in ribosome and tRNA synthesis, cell division (murC, mraY, ddl, bolA) and starvation control (rssB, fadB). The sizes of the Y. enterocolitica growth phase and temperature regulons show the importance of these signal in gene expression. Additional signals present during host infection lead to induction of further genes in vivo. However, the analysis of Y. pseudotuberculosis showed that growth in stationary phase at 37°C resembles infection conditions most (Fig. 3.3). This shows that the in vitro cultures used in this are a good system to mimic the conditions inside and outside host organisms.

105

5 Discussion

The task of ensuring nutrient availability is particularly challenging in the gut of host organisms. This is a highly individual environment, which depends a lot on the specific host, its microbiota and its diet. When looking at host adaptation of Yersinia, the anatomy of the host organism also needs to be taken into accounts. The gastrointestinal tract of potential host organisms varies significantly. For example, the diameter of the small intestine of humans (5 cm), pigs (2.5 – 3.5 cm) and rats (0.3 – 0.5 cm) is quite different (Kararli, 1995). Also, humans have a poorly defined cecal region. The porcine cecum is several orders of magnitude larger than that of humans (Kararli, 1995). Moreover, the log10 numbers of viable organisms per gram wet weight in the stomach and the proximal small intestine differ from 0 – 5 in mice to 7 – 9 in humans. In other areas of the gastrointestinal tract, the numbers are more similar, but variations still occur (Kararli, 1995). This demonstrates that within different host organisms, enteropathogenic Yersinia have to adapt differently to the gastrointestinal tract.

During the course of an infection, enteric pathogens encounter different situations with unique nutrient availabilities. The diet of the host, the reaction of the immune system and the composition of the microbiota shape the supply of nutrients. It was shown that the composition of metabolites varies in different parts of the human gut (Wang et al., 2007). The gut is rich in metabolites. However, it is not considered a nutrient rich, but rather a competitive environment for bacteria (Staib and Fuchs, 2014). The importance of nutrients for enteropathogens during an infection has been described and reviewed intensively (reviewed by Abu Kwaik and Bumann, 2013; Staib and Fuchs, 2014). In the gut, bacteria have to compete for available nutrients not only with the host itself, but also with the dense microbiota. This makes the host intestine a challenging environment for Yersinia. The competition between the commensal microbiota and the enteropathogens requires Yersinia to occupy a metabolic niche to ensure survival and proliferation.

It has been shown for several pathogens, that small changes in the gene expression pattern results in an advantage during infection. For example, S. enterica serotype Typhimurium uses the electron acceptor tetrathionate, which results in a growth advantage over the microbiota (Winter et al., 2010). The non-pathogenic E. coli strain Nissle 1917 was able to outcompete and reduce the colonization of Salmonella enterica in the mouse model (Deriu et al., 2013). Both bacteria acquire iron ions by a similar mechanism. However, E. coli Nissle

106

5 Discussion

1917 is able to avoid iron restriction by the host, gaining a colonization advantage over S. enterica (Deriu et al., 2013). Even within the same species there is competition for nutrients. Commensal E. coli strains can reduce or even eliminate the growth of a pathogenic strain in the intestine (Leatham et al., 2009). This is due to the fact that the commensal strains use five sugars that are important carbon sources for the pathogenic strain, inhibiting the pathogen by nutrient depletion (Maltby et al., 2013). Moreover, Legionella has been shown to promote proteasomal degradation of host proteins to generate amino acids which it can then use as a carbon and energy source (Price et al., 2011). Also, Salmonella exploits several nutrients in the host simultaneously to overcome the low amounts of given nutrients in the gut (Steeb et al., 2013). A similar mechanism can be observed for Y. enterocolitica and Y. pseudotuberculosis, which express a variety of nutrient uptake systems under all tested conditions (Datasets S3.4, S4.5, S4.6).

One aim of the present study was to investigate how different environmental conditions lead to changes in Yersinia gene expression. It has been shown before for Y. enterocolitica that small genetic or regulatory changes lead to significant changes in their gene expression pattern (Schaake et al., 2014; Uliczka et al., 2011). Previous findings were confirmed in this study, such as the general induction of the virulence plasmid at 37°C or the differential expression of the virulence factors rovA and invA between Y. enterolititca serotypes (Bölin et al., 1985; Straley and Perry, 1995; Uliczka et al., 2011). However, other interesting gene expression patters were found that could influence the outcome of an infection.

Carbohydrates A study comparing 18 human microbiomes identified 156 carbohydrate-active enzymes, including carbohydrate binding molecules, glycosyltransferases, polysaccharide lyases and carbohydrate-esterases (Turnbaugh et al., 2009). This shows that a lot of potential carbon sources are processed by microbiota. Finding the right niche of carbohydrate utilization is an important factor for survival of pathogens. This is supported by the presence of the aga operon only in the genome of YeO:3, which allows the bacteria to grow on N-acetyl- galactosamine (Batzilla et al., 2011). N-acetylgalactosamine is the major amino sugar in porcine mucin, which is in consistence with the ability of YeO:3 to colonize pigs in contrast to YeO:8 (Batzilla et al., 2011; Schaake et al., 2013). When it comes to the expression of carbohydrate processing enzymes, strong variations in the expression patterns were observed between different Yersinia isolates. The mannose

107

5 Discussion

uptake operon is strongly upregulated in YeO:3 compared to YeO:8 under all in vitro conditions, suggesting that mannose, fructose and sorbose are important carbohydrate sources for this strain. Contrary to this, maltose seems to be a more important carbohydrate source for YeO:8, as indicated by the upregulation of the malMKEFGlamB operon in this serotype. The fruBKA operon, encoding genes for the uptake of fructose, is strongly upregulated in Y. pseudotuberculosis during infection of the Peyer’s Patches. The respective deletion mutant is significantly impaired during an infection, underlining the importance of this particular carbon source for Y. pseudotuberculosis during infection. However, fruBKA is expressed only to a very low extend in Y. pseudertuberculosis under in vitro conditions. In Y. enterocolitica, in contrast, fruBKA is highly expressed already in vitro. For each of the investigated Y. entercolitica serotypes, one condition resulted in a very high expression of this operon. These observations show that the same operon can be differently expressed within one species, similar to what has been observed for E. coli (Maltby et al., 2013). In YeO:8 the strongest expression was observed at 25°C in stationary growth phase, while in YeO:3 the highest expression was observed at 25°C in exponential growth phase (Datasets S4.5 and S4.6). This observation might also hint to adaptation towards different environmental niches outside the host. Since very high expression in Y. enterocolitica was only observed at 25°C but not at 37°C, fruBKA might be important either in the early phase of an infection or in conditions outside a host. It was previously suggested that the metabolic properties (“substrate degradation”) not only have advantages in host organisms but also during proliferation in food (Staib and Fuchs, 2014). Interestingly, in the Y. enterocolitica in vitro studies, a significant upregulation of genes encoding carbohydrate processing enzymes was observed in both strains at 25°C when compared to 37°C at both growth phases (Datasets S4.5 and S4.6). This indicates that during an infection other signals, such as the presence of certain sugar compounds or host cell signals, are required to induce the necessary systems. In general, Y. enterocolitica is able to metabolize cellobiose, sucrose and inositol in contrast to Y. pseudotuberculosis and Y. pestis (Reuter et al., 2014). These additional metabolic activities could provide a growth advantage for Y. enterocolitica.

Nitrate In Y. pseudotuberculosis, growth at 37°C in stationary phase best reflects the infection conditions (Fig. 3.3A). One exemplary operon that is upregulated in Y. enterocolitica at this condition is napFDABC. This operon is significantly stronger expressed in YeO:3 compared

108

5 Discussion

to YeO:8. In Y. pseudotuberculosis the strongest expression is observed at 25°C in stationary growth phase. The high expression of this operon involved in nitrate reduction might be advantageous for YeO:3. Nitrate is present in the intestine at physiologically relevant concentrations and has been shown to be important for colonization of mice. E. coli mutants lacking the nitrate reductase had major colonization defects (Jones et al., 2007).

Taurine uptake and urease expression Taurine was described as the major organic solute in mammals with higher taurine levels found in the stomach mucosa compared to other gut regions (Huxtable, 1992; Wang et al 2007). Taurine was suggested to serve as a carbon and nitrogen source (Cook and Denger, 2006). The taurine uptake system tauABCD is upregulated in YeO:3 at 25°C in both growth phases compared to YeO:8. Induction of taurine metabolism prior to infection might be an advantage, as the bacteria expressing tauABCD already at 25°C are ready to metabolize taurine as soon as they enter the host. A similar mechanism is known for invasin, which is needed to bind to the host epithelial cells (Heroven et al., 2007). Expression of the taurine uptake system might give an advantage during the infection of humans and pigs. Due to the smaller stomach size, resulting in shorter residence time, this disadvantage might not play a role for YeO:8 in the mouse model. In humans, however, the bacteria stay in the stomach for over 1 h (Bornhorst and Paul Singh, 2014). Therefore, the metabolism of taurine could be one important factor that gives YeO:3 and advantage over YeO:8 when it comes to the colonization of pigs and humans. Additionally, there is hardly any expression of this operon in Y. pseudotubercuolsis, neither in vitro nor in vivo. A similar argumentation applies to urease expression. The ureABCDE operon is induced in YeO:8 at 25°C in stationary growth phase, but in YeO:3 it is highly expressed under all in vitro conditions tested in this study. In Y. pseudotuberculosis the urease genes are most strongly expressed at 37°C in exponential growth phase and only low expression can be observed in bacteria in the Peyer’s Patches, supporting the role of this operon during the early stage of an infection. With pH values as low as 1.5 to 2.5 the acidic environment of the mammalian stomach is a natural barrier against infections with food-borne pathogens.

Urease catalyzes the hydrolysis of urea to NH3 (ammonia) that is immediately protonated to + NH4 , therefore neutralizing the presence of protons to reduce acidity (Miller and Maier, 2014). This is an essential mechanism for the pathogenesis of H. pylori (Mobley et al., 1991). Together, the higher expression of ureABCE and tauABCD give YeO:3 a selective

109

5 Discussion

advantage that might lead to a more successful colonization of the porcine and human host. The stomach capacity of pigs (6-8 l) and humans (1-1.6 l) is larger than that or rodents (Kararli, 1995). This suggests a longer residence time in the porcine gut for enteric Yersinia. Therefore, urease could be most important during infection of porcine hosts and contributing to the ability of YeO:3 to colonize pigs in contrast to YeO:8.

Motility It has been shown before that Y. pseudotuberculosis and YeO:8 are flagellated and motile at 25°C, but not at 37°C. YeO:3 was found to be non-motile at both temperatures (Uliczka et al., 2011). In accordance with that, the present study revealed a strong induction of flagella protein-encoding operons in YeO:8 compared to YeO:3 (Dataset S4.7, S4.8). High flagella expression can be both, an advantage and a disadvantage, during an infection. For uropathogenic E. coli flagella are important to promote bacterial dissemination, but flagella are also recognized by the host immune system (Lüthje and Brauner, 2014). The flagellar proteins of C. jejuni have immunogenic effects (Yeh et al., 2013). Higher expression levels of flagella can be an advantage with regard to colonizing a host or reaching new residence sites. On the other hand, it could also be a disadvantage when they are recognized by the host immune system. YeO:3 is flagellated and motile immediately after isolation from the intestine, but loses motility when grown in vitro (Uliczka et al., 2011). This tightly controlled expression of the flagella might be an advantage with regard to energy consumption, but also with regard to recognition by the host immune system.

Transposases Interestingly, among the genes upregulated in YeO:3 compared to YeO:8 are several genes encoding transposases. It can be hypothesized that this might be the reason for genetic flexibility in this serotype. Transposases are necessary to transfer mobile genetic elements to other regions of the genome. The insertion of mobile elements can result in the silencing of a gene, but it can also result in the activation of gene expression (Glansdorff et al., 1981). In E. coli, the insertion of a mobile element resulted in the inactivation of AcrR, a repressor of acrAB, leading to increased resistance to fluoroquinolones (Jellen-Ritter and Kern, 2001). A transposase from Acinetobacter baumanii is even transferred into the host cell nucleus, which results in the downregulation of the E-cadherin gene (Moon et al., 2012). Considering these examples where transposases have positive effects on bacterial survival, it is possible that constant expression of transposases is beneficial to YeO:3. High expression of

110

5 Discussion

transposase genes will result in a high mutation rate, which would make the YeO:3 population stochastically better adaptable to continuously changing environmentals.

Amino acids For Y. enterocolitica serotypes differences in genes expression involved in amino acid transport and metabolism were detected. However, they could not be assigned to a certain pathway. A reason for this might be that the bacteria were grown in the same, rich medium. Limitation of certain amino acids in the host could lead to an increase in the expression of additional genes. In Clostridium difficile, expression of genes that are responsible for the uptake of amino acids, carbohydrates and fatty acids is upregulated during an infection (Fletcher et al., 2018). Additionally it has been shown that L. monocytogenes strains lacking oligopeptide transporters are attenuated in vivo (Borezee et al., 2000; Schauer et al., 2010), demonstrating the need for amino acid utilization. However, Y. enterocolitica and Y. pseudotuberculosis depend only on the presence of the aspartic family of amino acids, indicating that other nutrients might be more limiting for Yersinia during an infection (Brubaker, 1991).

Enterotoxin YstA This study revealed a strong up-regulation of the enterotoxin encoding gene ystA in YeO:3 compared to YeO:8. Expression of ystA is not clearly serotype specific, but rather isolate specific. However, all isolates with a high expression level of ystA belong to serotype O:3. The isolates showing a high ystA expression have all been isolated after 2007 and have been isolated because of strong phenotypes in patients during infection. It is highly likely that the YstA toxin is involved in the pathogenicity in these isolates, as the ystA gene is present only in pathogenic, but not in non-pathogenic strains of Y. enterocolitica (Delor et al., 1990) YstA has been used in several studies to identify pathogenic Y. enterocolitica strains (Delor et al., 1990; Ibrahim et al., 1997; Platt-Samoraj et al., 2006). Members of the same toxin family have been found in ETEC (So and McCarthy, 1980; Moseley et al., 1983), V. cholera (Arita et al., 1986), Citrobacter freundii (Guarino et al., 1987, 1989) and Klebsiella (Klipstein and Engert, 1976). Cross-species transfer has been suggested between E. coli and Y. pestis (Yamamoto and Taneike, 2000). However, ystA is not present in Y. pseudotuberculosis (Thoerner et al., 2003). The induction of ystA expression in stationary phase confirms the finding of a previous study (Mikulskis et al., 1994). In this work, it has also been suggested that the regulation of ystA

111

5 Discussion

occurs mainly at the transcriptional level, which is supported by the suggested influence of H–NS, YmoA and RovA on ystA expression, identified in the present study. The upregulation of ystA in combination with high constitutive expression of invasion factor Invasin could contribute to the increased toxicity of YeO:3 Y1.

In general, the adaption of metabolic genes and the upregulation of the TCA (Fig. 4.5) in YeO:8 at 25°C shows that the Y. enterocolitica serotypes have potentially adjusted to different ecological niches. YeO:8 seems to be more adapted to an environmental life style whereas YeO:3 is better adapted to mammalian hosts as shown by the upregulation of urease and the taurine uptake system as well as the fact that most genes upregulated in YeO:3 compared to YeO:8 are more abundant at 37°C (Fig. 4.4D, E). It was recently shown that the exchange of one nucleotide in the -10 region of the gene encoding the PgtE outer membrane protease (which is linked to the virulence of African S. typhimurium ST313) has a great impact on the pathogenicity of Salmonella enterica. This mutation lead to the emergence of a highly epidemic strain (Hammarlöf et al., 2018). In S. aureus the mutation of a single nucleotide was sufficient to alter the host-tropism of the strain from humans to rabbits (Viana et al., 2015). For Campylobacter the adaption to vitamin B5 synthesis was necessary for the adaption to cattle as a host, showing that the differential expression of a single gene can be of advantage for the colonization of specific hosts (Sheppard et al., 2013). Additionally, the formation of biofilms by Y. pseudotuberculosis was found to be strain specific (Joshua et al., 2003). Accordingly, it is reasonable to suggest that small changes in the gene expression profile also influences Yersinia pathogenicity. The results for Y. pseudotuberculosis showed that genes upregulated under in vivo compared to in vitro conditions were mainly stress response genes and genes encoding host-adapted metabolic functions. These functions depend greatly on the circumstances that the bacteria encounter in a specific situation and might be different for each individual host. Therefore, it is likely that Yersinia expresses some genes only when a certain environmental signal is detected. In contrast to that it was found that the classical virulence genes are mainly induced by temperature or growth phase, showing that they are generally needed for a successful infection. One additional thing that has to be kept in mind is that this study focused on the long-term adaption to a given condition. With regard to cold temperatures it was shown that different genes were induced if cold shock or long time adaption to low temperatures were compared

112

5 Discussion

(Bresolin et al., 2006a, 2006b). Therefore, it is also possible that other genes might be detected when heat shock is applied or samples are taken shortly after the temperature shift. The differences in the gene expression profiles of Y. enterocolitica and Y. pseudotuberculosis might lead to an advantage in a certain environment, as the mentioned examples of E. coli, Salmonella and Legionella show (Deriu et al., 2013; Maltby et al., 2013; Price et al., 2011; Steeb et al., 2013). In YeO:8 and YeO:3, the same virulence factors are involved in host cell binding, but differences in their expression profile in response to environmental signals led to significant differences in adhesion to and invasion of host cells (Uliczka et al., 2011). The additional expression differences within the Yersinia species shown here might influence the outcome of a host infection. However, the effects of specific operons need be investigated in further detail.

Loss and gain of certain genes shapes the adaptation of Yersinia (Batzilla et al., 2011; McNally et al., 2016; Reuter et al., 2014), but also different expression patterns of genes influence bacterial lifestyle and the outcome of an infection. Reprogramming of fitness and virulence traits can give bacteria advantage in certain environmental reservoirs and hosts. This study shows how even closely related bacteria such as YeO:8 and YeO:3 differ in their expression profile, adapting them to certain niches. The present study provides evidence for differential epidemiology of Y. enterocolitica and Y. pseudotuberculosis. Similarities and differences in the gene expression pattern of enteropathogenic Yersinia were detected, indicating factors that were involved in niche adaption of the analyzed strains. Both projects of this study provided a high-resolution data set enabling the identification of transcriptional changes in response to different stimuli. This offers a great source and valuable tool for continuous comparative studies on the gene expression, host adaption and regulatory features of enteropathogenic Yersinia. Although the effects of specific genes have to be further investigated, this study provides a useful basis for the analysis of the gene expression profile in response to environmental signals in enteropathogenic Yersinia. Moreover, the comparative transcriptional analysis provides evidence for differentially ecology, niche-adaption and epidemiology of Yersinia.

113

5 Discussion

5.4 References

Abu Kwaik, Y., and Bumann, D. (2013). Microbial quest for food in vivo: “nutritional virulence” as an emerging paradigm. Cell. Microbiol. 15, 882–890.

Arita, M., Takeda, T., Honda, T., and Miwatani, T. (1986). Purification and characterization of Vibrio cholerae non-O1 heat-stable enterotoxin. Infect. Immun. 52, 45–49.

Beauregard, A., Smith, E.A., Petrone, B.L., Singh, N., Karch, C., McDonough, K.A., and Wade, J.T. (2013). Identification and characterization of small RNAs in Yersinia pestis. RNA Biol 10,397–405.

Bohn, C., Rigoulay, C., Chabelskaya, S., Sharma, C.M., Marchais, A., Skorski, P., Borezée- Durant, E., Barbet, R., Jacquet, E., Jacq, A., Gautheret D., Felden B., Vogel J. and Bouloc P.(2010). Experimental discovery of small RNAs in Staphylococcus aureus reveals a riboregulator of central metabolism. Nucleic Acids Res. 38, 6620–6636.

Bölin, I., Portnoy, D.A., and Wolf-Watz, H. (1985). Expression of the temperature-inducible outer membrane proteins of yersiniae. Infect. Immun. 48, 234–240.

Borezee, E., Pellegrini, E., and Berche, P. (2000). OppA of Listeria monocytogenes, an oligopeptide-binding protein required for bacterial growth at low temperature and involved in intracellular survival. Infect. Immun. 68, 7069–7077.

Bornhorst, G.M., and Paul Singh, R. (2014). Gastric digestion in vivo and in vitro: how the structural aspects of food influence the digestion process. Annu Rev Food Sci Technol 5, 111–132.

Bresolin, G., Morgan, J.A.W., Ilgen, D., Scherer, S., and Fuchs, T.M. (2006a). Low temperature- induced insecticidal activity of Yersinia enterocolitica. Mol. Microbiol. 59, 503–512.

Bresolin, G., Neuhaus, K., Scherer, S., and Fuchs, T.M. (2006b). Transcriptional analysis of long- term adaptation of Yersinia enterocolitica to low-temperature growth. J. Bacteriol. 188, 2945–2958.

Brubaker, R.R. (1991). Factors promoting acute and chronic diseases caused by yersiniae. Clin. Microbiol. Rev. 4, 309–324.

Butcher, J., and Stintzi, A. (2013). The transcriptional landscape of Campylobacter jejuni under iron replete and iron limited growth conditions. PLoS ONE 8, e79475.

Cook, A.M., and Denger, K. (2006). Metabolism of taurine in microorganisms: a primer in molecular biodiversity? Adv. Exp. Med. Biol. 583, 3–13.

114

5 Discussion

Deriu, E., Liu, J.Z., Pezeshki, M., Edwards, R.A., Ochoa, R.J., Contreras, H., Libby, S.J., Fang, F.C., and Raffatellu, M. (2013). Probiotic bacteria reduce salmonella typhimurium intestinal colonization by competing for iron. Cell Host Microbe 14, 26–37.

Dugar, G., Herbig, A., Förstner, K.U., Heidrich, N., Reinhardt, R., Nieselt, K., and Sharma, C.M. (2013). High-resolution transcriptome maps reveal strain-specific regulatory features of multiple Campylobacter jejuni isolates. PLoS Genet. 9, e1003495.

Fletcher, J.R., Erwin, S., Lanzas, C., and Theriot, C.M. (2018). Shifts in the Gut Metabolome and Clostridium difficile Transcriptome throughout Colonization and Infection in a Mouse Model.MSphere3.

Geissmann, T., Chevalier, C., Cros, M.-J., Boisset, S., Fechter, P., Noirot, C., Schrenzel, J., François, P., Vandenesch, F., Gaspin, C. and Romby P. (2009). A search for small noncoding RNAs in Staphylococcus aureus reveals a conserved sequence motif for regulation. Nucleic Acids Res. 37, 7239–7257.

Geng, J., Song, Y., Yang, L., Feng, Y., Qiu, Y., Li, G., Guo, J., Bi, Y., Qu, Y., Wang, W., Wang X., Guo, Z. Yang, R. and Han, Y. (2009). Involvement of the post-transcriptional regulator Hfq in Yersinia pestis virulence. PLoS ONE 4, e6213.

Georg, J., and Hess, W.R. (2011). cis-antisense RNA, another level of gene regulation in bacteria. Microbiol. Mol. Biol. Rev. 75, 286–300.

Glansdorff, N., Charlier, D., Zafarullah, M. (1981). Activation of gene expression by IS2 and IS3. Cold Spring Harb Symp Quant Biol. 45, 153-6.

Guarino, A., Capano, G., Malamisura, B., Alessio, M., Guandalini, S., and Rubino, A. (1987). Production of Escherichia coli STa-like heat-stable enterotoxin by Citrobacter freundii isolated from humans. J. Clin. Microbiol. 25, 110–114.

Guarino, A., Giannella, R., and Thompson, M.R. (1989). Citrobacter freundii produces an 18-amino- acid heat-stable enterotoxin identical to the 18-amino-acid Escherichia coli heat-stable enterotoxin (ST Ia). Infect. Immun. 57, 649–652.

Hammarlöf, D.L., Kröger, C., Owen, S.V., Canals, R., Lacharme-Lora, L., Wenner, N., Schager, A.E., Wells, T.J., Henderson, I.R., Wigley, P., Hokamp K., Feasey N., Gordon M.A. and Hinton J.C.D. (2018). Role of a single noncoding nucleotide in the evolution of an epidemic African clade of Salmonella. Proc. Natl. Acad. Sci. U.S.A. 115, E2614–E2623.

Henkin, T.M. (2008). Riboswitch RNAs: using RNA to sense cellular metabolism. Genes Dev. 22, 3383–3390.

115

5 Discussion

Heroven, A.K., and Dersch, P. (2014). Coregulation of host-adapted metabolism and virulence by pathogenic yersiniae. Front Cell Infect Microbiol 4, 146.

Heroven, A.K., Böhme, K., Tran-Winkler, H., and Dersch, P. (2007). Regulatory elements implicated in the environmental control of invasin expression in enteropathogenic Yersinia. Adv. Exp. Med. Biol. 603, 156–166.

Holmqvist, E., Li, L., Bischler, T., Barquist, L., and Vogel, J. (2018). Global Maps of ProQ Binding In Vivo Reveal Target Recognition via RNA Structure and Stability Control at mRNA 3’ Ends. Mol. Cell 70, 971-982.e6.

Huxtable, R.J. (1992). Physiological actions of taurine. Physiol. Rev. 72, 101–163.

Ibrahim, A., Liesack, W., Griffiths, M.W., and Robins-Browne, R.M. (1997). Development of a highly specific assay for rapid identification of pathogenic strains of Yersinia enterocolitica based on PCR amplification of the Yersinia heat-stable enterotoxin gene (yst). J. Clin. Microbiol. 35, 1636–1638.

Jellen-Ritter, A.S., and Kern, W.V. (2001). Enhanced expression of the multidrug efflux pumps AcrAB and AcrEF associated with insertion element transposition in Escherichia coli mutants Selected with a fluoroquinolone. Antimicrob. Agents Chemother. 45, 1467–1472.

Jones, S.A., Chowdhury, F.Z., Fabich, A.J., Anderson, A., Schreiner, D.M., House, A.L., Autieri, S.M., Leatham, M.P., Lins, J.J., Jorgensen, M., Cohen P.S., and Conway T. (2007). Respiration of Escherichia coli in the mouse intestine. Infect. Immun. 75, 4891–4899.

Joshua, G.W.P., Karlyshev, A.V., Smith, M.P., Isherwood, K.E., Titball, R.W., and Wren, B.W. (2003). A Caenorhabditis elegans model of Yersinia infection: biofilm formation on a biotic surface. Microbiology (Reading, Engl.) 149, 3221–3229.

Kakoschke, T., Kakoschke, S., Magistro, G., Schubert, S., Borath, M., Heesemann, J., and Rossier, O. (2014). The RNA chaperone Hfq impacts growth, metabolism and production of virulence factors in Yersinia enterocolitica. PLoS ONE 9, e86113.

Kararli, T.T. (1995). Comparison of the gastrointestinal anatomy, physiology, and biochemistry of humans and commonly used laboratory animals. Biopharm Drug Dispos 16, 351–380.

Klipstein, F.A., and Engert, R.F. (1976). Purification and properties of Klebsiella pneumoniae heat- stable enterotoxin. Infect. Immun. 13, 373–381.

Kopf, M., and Hess, W.R. (2015). Regulatory RNAs in photosynthetic cyanobacteria. FEMS Microbiol. Rev. 39, 301–315.

116

5 Discussion

Kortmann, J., and Narberhaus, F. (2012). Bacterial RNA thermometers: molecular zippers and switches. Nat. Rev. Microbiol. 10, 255–265.

Kröger, C., Dillon, S.C., Cameron, A.D.S., Papenfort, K., Sivasankaran, S.K., Hokamp, K., Chao, Y., Sittka, A., Hébrard, M., Händler, K., Colgan A., Leekitcharoenphon P., Langridge GC., Lohan A.J., Loftus, B., Lucchini S., Ussery D.W., Dorman C.J., Thomson N.R., Vogel, J. and Hinton J.C. (2012). The transcriptional landscape and small RNAs of Salmonella enterica serovar Typhimurium. Proc. Natl. Acad. Sci. U.S.A. 109, E1277-1286.

Kröger, C., Colgan, A., Srikumar, S., Händler, K., Sivasankaran, S.K., Hammarlöf, D.L., Canals, R., Grissom, J.E., Conway, T., Hokamp, K. and Hinton J.C. (2013). An infection-relevant transcriptomic compendium for Salmonella enterica Serovar Typhimurium. Cell Host Microbe 14, 683– 695.

Lasa, I., Toledo-Arana, A., Dobin, A., Villanueva, M., de los Mozos, I.R., Vergara-Irigaray, M., Segura, V., Fagegaltier, D., Penadés, J.R., Valle, J., Solano, C. and Gingeras, T.R. (2011). Genome-wide antisense transcription drives mRNA processing in bacteria. Proc. Natl. Acad. Sci. U.S.A. 108, 20172–20177.

Leatham, M.P., Banerjee, S., Autieri, S.M., Mercado-Lubo, R., Conway, T., and Cohen, P.S. (2009). Precolonized human commensal Escherichia coli strains serve as a barrier to E. coli O157:H7 growth in the streptomycin-treated mouse intestine. Infect. Immun. 77, 2876–2886.

Lüthje, P., and Brauner, A. (2014). Virulence factors of uropathogenic E. coli and their interaction with the host. Adv. Microb. Physiol. 65, 337–372.

Maltby, R., Leatham-Jensen, M.P., Gibson, T., Cohen, P.S., and Conway, T. (2013). Nutritional basis for colonization resistance by human commensal Escherichia coli strains HS and Nissle 1917 against E. coli O157:H7 in the mouse intestine. PLoS ONE 8, e53957.

McNally, A., Thomson, N.R., Reuter, S., and Wren, B.W. (2016). “Add, stir and reduce”: Yersinia spp. as model bacteria for pathogen evolution. Nat. Rev. Microbiol. 14, 177–190.

Mendoza-Vargas, A., Olvera, L., Olvera, M., Grande, R., Vega-Alvarado, L., Taboada, B., Jimenez-Jacinto, V., Salgado, H., Juárez, K., Contreras-Moreira, B., Huerta, A.M., Collado-Vides, J. and Morett E. (2009). Genome-wide identification of transcription start sites, promoters and transcription factor binding sites in E. coli. PLoS ONE 4, e7526.

Miller, E.F., and Maier, R.J. (2014). Ammonium metabolism enzymes aid Helicobacter pylori acid resistance. J. Bacteriol. 196, 3074–3081.

117

5 Discussion

Mobley, H.L., Hu, L.T., and Foxal, P.A. (1991). Helicobacter pylori urease: properties and role in pathogenesis. Scand. J. Gastroenterol. Suppl. 187, 39–46.

Moon, D.C., Choi, C.H., Lee, S.M., Lee, J.H., Kim, S.I., Kim, D.S., and Lee, J.C. (2012). Nuclear translocation of Acinetobacter baumannii transposase induces DNA methylation of CpG regions in the promoters of E-cadherin gene. PLoS ONE 7, e38974.

Nakao, H., Watanabe, H., Nakayama, S., and Takeda, T. (1995). yst gene expression in Yersinia enterocolitica is positively regulated by a chromosomal region that is highly homologous to Escherichia coli host factor 1 gene (hfq). Mol. Microbiol. 18, 859–865.

Nicolas, P., Mäder, U., Dervyn, E., Rochat, T., Leduc, A., Pigeonneau, N., Bidnenko, E., Marchadier, E., Hoebeke, M., Aymerich, S., et al. (2012). Condition-dependent transcriptome reveals high-level regulatory architecture in Bacillus subtilis. Science 335, 1103–1106.

Pal, C., Papp, B. and Lercher M.J. (2005). Adaptive evolution of bacterial metabolic networks by horizontal gene transfer. Nat Genet 37, 1372-1375

Petersen, L., Larsen, T.S., Ussery, D.W., On, S.L.W., and Krogh, A. (2003). RpoD promoters in Campylobacter jejuni exhibit a strong periodic signal instead of a -35 box. J. Mol. Biol.326,1361–1372.

Platt-Samoraj, A., Ugorski, M., Szweda, W., Szczerba-Turek, A., Wojciech, K., and Procajło, Z. (2006). Analysis of the presence of ail, ystA and ystB genes in Yersinia enterocolitica strains isolated from aborting sows and aborted fetuses. J. Vet. Med. B Infect. Dis. Vet. Public Health 53, 341–346.

Price, C.T.D., Al-Quadan, T., Santic, M., Rosenshine, I., and Abu Kwaik, Y. (2011). Host proteasomal degradation generates amino acids essential for intracellular bacterial growth. Science 334, 1553–1557.

Qiu, Y., Cho, B.-K., Park, Y.S., Lovley, D., Palsson, B.Ø., and Zengler, K. (2010). Structural and operational complexity of the Geobacter sulfurreducens genome. Genome Res. 20, 1304–1311.

Raghavan, R., Sloan, D.B., and Ochman, H. (2012). Antisense transcription is pervasive but rarely conserved in enteric bacteria. MBio 3.

118

5 Discussion

Sakurai, I., Stazic, D., Eisenhut, M., Vuorio, E., Steglich, C., Hess, W.R., and Aro, E.-M. (2012). Positive regulation of psbA gene expression by cis-encoded antisense RNAs in Synechocystis sp. PCC 6803. Plant Physiol. 160, 1000–1010.

Schauer, K., Geginat, G., Liang, C., Goebel, W., Dandekar, T., and Fuchs, T.M. (2010). Deciphering the intracellular metabolism of Listeria monocytogenes by mutant screening and modelling. BMC Genomics 11, 573.

Schiano, C.A., Bellows, L.E., and Lathem, W.W. (2010). The small RNA chaperone Hfq is required for the virulence of Yersinia pseudotuberculosis. Infect. Immun. 78, 2034–2044.

Soutourina, O.A., Monot, M., Boudry, P., Saujet, L., Pichon, C., Sismeiro, O., Semenova, E., Severinov, K., Le Bouguenec, C., Coppée, J.-Y., Dupuy, B. and Martin-Verstraete, I. (2013). Genome-wide identification of regulatory RNAs in the human pathogen Clostridium difficile. PLoS Genet. 9, e1003493.

Staib, L., and Fuchs, T.M. (2014). From food to cell: nutrient exploitation strategies of enteropathogens. Microbiology (Reading, Engl.) 160, 1020–1039.

Stazic, D., and Voß, B. (2016). The complexity of bacterial transcriptomes. J. Biotechnol. 232, 69–78.

Steeb, B., Claudi, B., Burton, N.A., Tienz, P., Schmidt, A., Farhan, H., Mazé, A., and Bumann, D. (2013). Parallel exploitation of diverse host nutrients enhances Salmonella virulence. PLoS Pathog. 9, e1003301.

Storz, G., Vogel, J., and Wassarman, K.M. (2011). Regulation by small RNAs in bacteria: expanding frontiers. Mol. Cell 43, 880–891.

Straley, S.C., and Perry, R.D. (1995). Environmental modulation of gene expression and pathogenesis in Yersinia. Trends Microbiol. 3, 310–317.

119

5 Discussion

Thoerner, P., Bin Kingombe, C.I., Bögli-Stuber, K., Bissig-Choisat, B., Wassenaar, T.M., Frey, J., and Jemmi, T. (2003). PCR detection of virulence genes in Yersinia enterocolitica and Yersinia pseudotuberculosis and investigation of virulence gene distribution. Appl. Environ. Microbiol. 69, 1810–1816.

Thomason, M.K., Bischler, T., Eisenbart, S.K., Förstner, K.U., Zhang, A., Herbig, A., Nieselt, K., Sharma, C.M., and Storz, G. (2015). Global transcriptional start site mapping using differential RNA sequencing reveals novel antisense RNAs in Escherichia coli. J. Bacteriol. 197, 18–28.

Turnbaugh, P.J., Hamady, M., Yatsunenko, T., Cantarel, B.L., Duncan, A., Ley, R.E., Sogin, M.L., Jones, W.J., Roe, B.A., Affourtit, J.P., Egholm, M., Henrissat, B., Heath, A.C., Knight, R. and Gordon, J.I. (2009). A core gut microbiome in obese and lean twins. Nature 22, 480-4.

Uliczka, F., Pisano, F., Schaake, J., Stolz, T., Rohde, M., Fruth, A., Strauch, E., Skurnik, M., Batzilla, J., Rakin, A., Heesemann, J. and Dersch, P. (2011). Unique cell adhesion and invasion properties of Yersinia enterocolitica O:3, the most frequent cause of human Yersiniosis. PLoS Pathog. 7, e1002117.

Viana, D., Comos, M., McAdam, P.R., Ward, M.J., Selva, L., Guinane, C.M., González-Muñoz, B.M., Tristan, A., Foster, S.J., Fitzgerald, J.R., and Penadés J.R. (2015). A single natural nucleotide mutation alters bacterial pathogen host tropism. Nat. Genet. 47, 361–366.

Wang, X., Su, M., Qiu, Y., Ni, Y., Zhao, T., Zhou, M., Zhao, A., Yang, S., Zhao, L. and Jia, W. (2007). Metabolic regulatory network alterations in response to acute cold stress and ginsenoside intervention. J Proteome Res. 6, 3449-3455.

Waters, L.S., and Storz, G. (2009). Regulatory RNAs in bacteria. Cell 136, 615–628.

Winkler, W.C., and Breaker, R.R. (2005). Regulation of bacterial gene expression by riboswitches. Annu. Rev. Microbiol. 59, 487–517.

Winter, S.E., Thiennimitr, P., Winter, M.G., Butler, B.P., Huseby, D.L., Crawford, R.W., Russell, J.M., Bevins, C.L., Adams, L.G., Tsolis, R.M., Roth J.R. Bäumler A.J. (2010). Gut inflammation provides a respiratory electron acceptor for Salmonella. Nature 467, 426–429.

Yamamoto, T., and Taneike, I. (2000). The sequences of enterohemorrhagic Escherichia coli and Yersinia pestis that are homologous to the enteroaggregative E. coli heat-stable enterotoxin gene: cross-species transfer in evolution. FEBS Lett. 472, 22–26.

Yeh, H.-Y., Hiett, K.L., Line, J.E., Oakley, B.B., and Seal, B.S. (2013). Construction, expression, purification and antigenicity of recombinant Campylobacter jejuni flagellar proteins. Microbiol. Res. 168, 192–198.

120

6 Outlook

In this study it was demonstrated that the expression of ystA greatly differs between Y. enterocolitca isolates. The function of the enterotoxin YstA for Yersinia pathogenesis should be further characterized future studies. The effect of purified YstA on various porcine and human cell types could be characterized to get a deeper knowledge of the YstA function. To clarify the importance of YstA for Yersina virulence, a deletion mutant should be tested in in vitro and especially in in vivo experiments. It was shown that H-NS and RovA are involved in the regulation of ystA expression. In future studies, this effect should be investigated further. Binding studies with ystA DNA should be performed to analyze if the effect of the regulators is direct or indirect. Moreover, the influence of further regulators should be investigated.

In general, the present study provides a global map of transcriptional start sites of Yersinia. Performing RNA-sequencing under different conditions such as acid stress, oxidative stress or anaerobic growth could add further TSS to the TSS map presented in this study.

The effect of genes differentially expressed between the two tested strains of Y. enterocolitica should be further investigated. Deletion mutants could be tested in the mouse or minipig model to determine which genes actually have an influence during infections. Since pigs are the main source of infections for humans, the minipig model is a useful tool to find out which genes that are involved in the colonization of this host organism. Of special interest would be to determine if the upregulation of genes that were induced in YeO:3 would lead to the ability of YeO:8 to colonize porcine hosts.

For both Yersinia species, Y. enterocolitica and Y. pseudotuberculosis, the function of several of the newly identified sRNAs should be investigated to find out whether any of them plays a role in Yersinia virulence. For example, the binding of Hfq to the newly identified sRNA candidates could be tested. Moreover, deletion mutants should be tested for phenotypic differences.

The existing data sets could be used to perform DEseq analysis on the core proteome of enteropathogenic Yersinia species to detect which genes are differentially expressed between Y. enterocolitica and Y. pseudotuberculosis on a global scale to further analyze the specific adaptation of the bacteria with regards to differential ecology and epidemiology.

121

6 Outlook

122

7 Appendix

7.1 Curriculum Vitae

Personal Details

Name Carina Maria Schmühl

Date of birth 8th February 1990 Degree Master of Science Nationality German Address Molenberger Straße 38114 Braunschweig Phone +49 176 84427538 E-mail [email protected]

Education

Since Oct 2014 Helmholtz Centre for Infection Research Major Topics Infection Biology, Molecular Biology, Transcriptomics Dissertation Mechanisms of niche adaption of Yersinia

Oct 2012 – Sept 2014 RWTH Aachen University Course Molecular and Applied Biotechnology Master of Science (Grade 1.3) Major Topics Medical biotechnology, microbiology Master Thesis Influence of environmental signals on virulence gene expression in Yersinia pseudotuberculosis (HZI, Braunschweig)

Oct 2009 – Sept 2012 RWTH Aachen University Course Biotechnology / Molecular Biotechnology Bachelor of Science (Grade 1.6) Bachelor Thesis Functional studies of four uncharacterised coiled coil proteins with putative roles in cell architecture of Streptomyces coelicolor (Lunds University, Lund, Sweden)

Aug 2000 – Jun 2009 Gymnasium der Gemeinde Kreuzau A- level: 1.1

123

7 Appendix

7.3 Acknowledgment

Ich bedanke mich bei Prof. Dr. Petra Dersch für die Möglichkeit, in ihrer Gruppe an diesem interessanten Projekt zu Arbeiten, für ihre Unterstützung und Motivation, auch in schwierigeren Phasen des Projektes.

Prof. Dr. Peter Valentin-Weigand danke ich für die Übernahme des zweiten Gutachtens, sowie die kontinuierliche Begleitung im Rahmen meiner Betreuungsgruppe und meiner Prüfungskommission. PD Dr. Simone Bergmann danke ich ebenfalls für die Begleitung als Teil meiner Betreuungsgruppe und meiner Prüfungskommission.

Des weiteren möchte ich mich bei der HGNI Graduate School der TiHo Hannover für die finanzielle Unterstützung bedanken. Außerdem bedanke ich mich bei den Mitarbeiterinnen der TiHo, die immer hilfsbereit und freundlich waren und sich Zeit für meine Fragen genommen haben.

Ein großes Danke schön geht an unsere PostDocs Kathi, Sabrina und Aaron, für konstante Hilfe jeglicher Art: Aaron für das Einarbeiten in die Welt des RNA-sequencing; Kathi dafür, dass sie mich „adoptiert“ und immer unterstützt hat; Sabrina für die vielen tollen Gespräche über diverse Themen, die mal hilfreich, mal motivierend und mal einfach unterhaltsam waren. Vielen Dank an unsere tollen TAs – Bettina, Tanja und Sandra – für ihre große Hilfsbereitschaft und nette Gesellschaft Bei meinen Mit-Doktoranden – Vanessa, Ines, Maria, Marcel, Paweena– bedanke ich mich für eine tolle Atmosphäre innerhalb und außerhalb der Arbeit. Und für die vielen Kekse! Danke auch an unsere gute Seele Claudia, die uns den Rücken frei hält und uns so einiges erleichtert. Den Ex-MIBIS – Jörn, Jessica, Franzi, Maik, und Wiebke – danke ich für die Hilfe, die guten Ratschläge und viele lustige Momente.

Ein großes Danke an meine Mädels zu hause – Caro, Anke und Christina – dafür, dass ihr mir immer zugehört habt, mich motiviert habt und für Spaß und Ablenkung gesorgt habt. Riesigen Dank an meine Eltern und Michi für die konstante Unterstützung. Danke, dass ihr so viel Verständnis hattet und mich immer nach Kräften unterstützt habt. Ein ganz besonderer Dank geht an Niclas, der mir in der ganzen Zeit eine wichtige Stütze war. Danke, dass du mein bester Freund bist, mich mit leckerem Essen versorgst, mich immer zu lachen bringst und mir in allen Momenten Rückhalt gegeben hast. Danke, dass ich immer auf euch zählen kann!

124